Advisory Committee on Fisheries Management ICES CM 2001/ACFM:03

REPORT OF THE

EIFAC/ICES WORKING GROUP ON EELS

St. Andrews, N.B., Canada 28 August - 1 September 2000

This report is not to be quoted without prior consultation with the General Secretary. The document is a report of an expert group under the auspices of the International Council for the Exploration of the Sea and does not necessarily represent the views of the Council.

International Council for the Exploration of the Sea Conseil International pour l’Exploration de la Mer

Palægade 2–4 DK–1261 Copenhagen K Denmark TABLE OF CONTENTS Section Page 1 INTRODUCTION...... 1 1.1 Main Tasks ...... 1 1.2 Participants...... 1 2 TRENDS IN RECRUITMENT AND FISHERIES, CAUSES AND EFFECTS ON STOCK AND YIELD ...... 2 2.1 Trends in Commercial Landings of American Eels...... 2 2.1.1 Reported elver catches...... 2 2.1.2 Reported yellow and silver eel catches...... 2 2.1.3 Reliability of reported catches...... 3 2.2 Trends in Abundance Indices ...... 3 2.2.1 Recruitment stages ...... 3 2.2.2 Yellow and silver eel stages ...... 3 2.3 Causes of Observed Trends in Recruitment ...... 5 2.3.1 Oceanic factors...... 5 2.3.2 Effects on continental life stages...... 6 2.3.2.1 Continental factors affecting the incoming recruitment ...... 6 2.3.2.2 Continental factors affecting the fecundity of the spawning stock...... 6 2.4 Summary/Conclusions...... 6 Tables 2.1.1.1-2.2.2.1 ...... 8 Figures 2.1.2.1.-2.31.5...... 18 3 IMPACT OF FISHERIES IN SELECTED SYSTEMS ON EGG PRODUCTION...... 34 3.1 Impacts of Fisheries on Escapement...... 34 3.1.1 Elver fisheries...... 34 3.1.2 Yellow/silver eel fisheries...... 34 3.1.3 Silver eels ...... 35 3.1.4 Natural mortality ...... 35 3.2 Fish Passage ...... 36 3.2.1 Potential effects of dams on migration...... 36 3.2.2 Habitat change...... 37 3.2.3 Upstream fish passage facilities ...... 37 3.2.4 Monitoring of juvenile recruitment ...... 37 3.2.5 Downstream passage ...... 38 3.2.6 Facilities and management action needed to ensure sustainable fisheries...... 38 3.3 Other Anthropogenic Effects...... 38 3.3.1 Physical habitat change ...... 38 3.3.2 Chemical habitat change ...... 39 3.3.2.1 Contaminants...... 39 3.3.2.2 Productivity ...... 39 3.3.3 Biological habitat change ...... 39 3.3.3.1 Introductions...... 39 3.3.3.2 Changes to spawning habitat ...... 40 Tables 3.1.1.1-3.2.1.1...... 41 Figures 3.1.2.2-3.1.4.1...... 44 4 OPTIONS FOR DEVELOPING ESCAPEMENT TARGETS FOR SELECTED SYSTEMS ...... 46 4.1 Evaluation of Human and Natural Impacts on American Eel Recruitment...... 46 4.1.1 The Paulik diagram ...... 46 4.1.2 Evaluation of impacts...... 48 4.1.3 Spawning escapement and the Precautionary Approach – life history insights...... 48 4.2 Biological Reference Points and the Precautionary Approach ...... 48 4.2.1 Reference points...... 48 4.2.2 Consequences of uncertainty...... 50 4.2.3 Proposed limit reference points in data-poor conditions ...... 50 4.2.4 Preliminary reference points for American eel...... 51 4.2.5 Limitations of Flim method...... 51 4.2.6 Application of Flim and Fpa...... 52 Tables 4.1.1-4.2.11...... 53 Figures 4.1.1-4.2.4.3...... 55 5 MANAGEMENT UNITS FOR ESCAPEMENT TARGETS ...... 64 5.1 Application of the Precautionary Approach ...... 64 5.2 Management Units...... 64

i Section Page

5.3 Multi-Jurisdictional Eel Stocks ...... 64 5.4 Outside of Canadian/US Jurisdiction ...... 64 6 MANAGEMENT ACTIONS FOR REQUIRED ESCAPEMENTS ...... 65 6.1 Current Management Plans ...... 65 6.2 Management Options...... 66 6.2.1 Measures to limit exploitation by fisheries...... 66 6.2.2 Upstream and downstream passage...... 67 6.2.3 Transfers and re-stocking ...... 67 6.3 Conclusions/ Recommendations...... 68 Table 6.1.1...... 69 7 COORDINATION OF MONITORING AND RESEARCH...... 74 7.1 Monitoring...... 74 7.1.1 Current monitoring: recruiting life stages ...... 74 7.1.2 Current monitoring: yellow eel ...... 74 7.1.3 Current monitoring: silver eel ...... 74 7.2 Research ...... 75 7.2.1 International coordination of research...... 75 7.2.2 Research priorities...... 75 7.2.2.1 Baseline landings and survey ...... 75 7.2.2.2 Ecological evaluation of biological processes...... 75 7.2.2.3 Research into anthropogenic impacts ...... 76 7.2.2.4 Research of basic methodological processes...... 76 7.3 International Coordination of Management...... 76 8 CONCLUSIONS AND RECOMMENDATIONS ...... 77 APPENDIX 1...... 78 APPENDIX 2...... 80 APPENDIX 3...... 84 APPENDIX 4...... 85 APPENDIX 5...... 86 @#

ii 1 INTRODUCTION

1.1 Main Tasks

At its 1999 Statutory Meeting, ICES resolved (C. Res. 1999/2ACFM13) that the EIFAC/ICES Working Group on Eels [WGEEL](Chair: Dr L. Marshall, Canada) will meet in St. Andrews NB, Canada from 28 August to 1 September, in 2000 to consider questions posed to ICES by the governments of Canada and USA. The terms of reference and sections of the report, in which the answers are provided, follow.

Terms of Reference Section a) assess trends in recruitment and their causes, in fisheries and the effects on stock and yield of the 2, 3 American eel; b) investigate the impact of fisheries on American eel in selected systems, especially with regard to 3 the consequences for spawner escapement; c) investigate the options for developing escapement targets for American eel for selected systems; 4 d) define relevant units where these targets would apply; 5 e) suggest type of management actions that may lead to the required escapement of American eel; 6, 8 f) advise on international coordination of research on American eel in the future. 7

The Working Group considered 25 Working Documents (Appendix 1) submitted by 21 participants; other references cited in the report are given in Appendix 2. Complete addresses for five authors who provided information used by the Working Group that was not referenceable due to previous publishing commitments, appear in Appendix 3. Definitions of key life history terms and fishing gear for the American eel, Anguilla rostrata, are provided in Appendix 4.

1.2 Participants

Boubée, J New Zealand Cairns, D. Canada Caron, F. Canada Casselman, J. Canada Castonguay, M. Canada Chaput, G. Canada Dekker, W Netherlands Jessop, B. Canada Marshall, L. Canada (Chair) Mathers, A Canada McCleave, J. USA Meerburg, D. Canada Ritter, J. Canada Peterson, R Canada Rosell, R. UK (NI) Secor, D. USA Sparholt, H. ICES staff Todd, P. New Zealand Vecchio, V. USA Weeder, J. USA Westerberg, H. Sweden

A full address list for the participants is provided in Appendix 5.

1 2 TRENDS IN RECRUITMENT AND FISHERIES, CAUSES AND EFFECTS ON STOCK AND YIELD

2.1 Trends in Commercial Landings of American Eels

2.1.1 Reported elver catches

Canada: Commercial harvests of American eel elvers began in the Bay of Fundy and Atlantic coastal areas of Nova Scotia and New Brunswick (Scotia-Fundy area) in 1989 (Table 2.1.1.1). Landings steadily increased to a peak of 4,122 kg in 1997 and then, because of reduced market demand, declined to 478 kg in 1999 and 622 kg in 2000.

United States: An elver fishery occurred in Maine during the late 1970s and collapsed due to poor market conditions (Table 2.1.1.1). It continued at a low level until the early 1990s when Asian demand for elvers for aquaculture greatly increased. The Asian market collapsed in 1999, resulting in much reduced fishing effort in all areas. The collapse resulted from decreased demand for elvers of North American origin consequent to high catches of Japanese eel elvers and the production of market-sized cultured eels in excess of Asian demand. Several other Atlantic coastal states also developed elver fisheries during the 1990s, but no catch data are available (Jessop 1997).

Caribbean and Central America: The periodic reporting of “river eel” catches in Caribbean and Central American countries by FAO (Table 2.1.1.2) are believed to be glass eels/elvers caught for export. Information available in FAO fisheries databases suggests catches of as much as 49 t of eel in some recent years by some countries, at prices which suggest most of the catch must be glass eels. However, incidental but well documented records of an older date (Fernandez and Vasquez 1978) quantify 0.4 to 7 t of glass eels being caught in some rivers in Holguin Province in Cuba. This suggests that catches of glass eel/elver from the Caribbean and Central American countries might well be underreported by FAO, and probably comprise a large share of glass eel catches of American eel. A better understanding and better recording of eel catches in Caribbean and Central/ South American countries is urgently required.

2.1.2 Reported yellow and silver eel catches

North America: Total reported landings in North America (Canada and USA), 1950-1998 peaked at about 2,900 t in 1975 (Figure 2.1.2.1). Catches declined thereafter to about 1,000 t in 1997-1998, a level last seen in the mid 1950s and early 1960s. Over the entire period the average reported catch of each of Canada and the USA approximated 800 t.

Canada: Reported yellow and silver eel catches began in 1884 and peaked firstly at nearly 1,150 t in the 1930s and secondly at 1,359 t in the late 1960s to early 1980s (Figure 2.1.2.2; Table 2.1.2.1). Catches in 1999 are the lowest in 49 years. Catches from Ontario and Québec comprised the majority of Canadian landings until the 1960s when landings from the southern Gulf of St. Lawrence became significant (Figure 2.1.2.2). Catches in Newfoundland and the Scotia- Fundy portion of Nova Scotia and New Brunswick only became significant portions of total Canadian landings in the last two decades (Table 2.1.2.1).

Yellow and silver eel catches in the St. Lawrence River/ Lake Ontario system have steadily declined since 1992 from approximately 600 t to 200 t in 1997 (Figure 2.1.2.3; Table 2.1.2.1). A partial closure of the Lake Ontario fishery due to contaminants occurred in 1998 and 1999, which may account for a minor fraction of the decline in those years. A longer time series of yellow eel catches from Lake Ontario and the upper St. Lawrence River indicates an earlier catch cycle (Figure 2.1.2.4). Landings in the New Brunswick and Prince Edward Island portions of the Gulf of St. Lawrence have also exhibited a four-fold decline over the past decade (Figs. 2.1.2.5, 6; Table 2.1.2.1). However, landings from the Nova Scotia region of the Gulf of St. Lawrence (Figure 2.1.2.6; Table 2.1.2.1) and from Newfoundland (Figure 2.1.2.5; Table 2.1.2.1) show some variability but no clear trend. By contrast, an upward trend is clearly apparent in catches south of the Gulf of St. Lawrence, in the Canadian Atlantic / Bay of Fundy regions. Catches from the Atlantic and Bay of Fundy portions of Nova Scotia, and from the Bay of Fundy portion of New Brunswick have increased about threefold since the mid or late 1980s (Figure 2.1.2.6; Table 2.1.2.1).

United States: The United States fishery for American eels extends from Maine to the Gulf of Mexico (Table 2.1.2.2; Figs 2.1.2.1, 7). Different geographic regions (north, Figure 2.1.2.8, middle, Figure 2.1.2.9, and south Atlantic, Figure 2.1.2.10, and Gulf of Mexico, Figure 2.1.2.11) exhibit differing time trends in, and magnitudes of, their fisheries. This reflects the differing nature of their fisheries and regional stock abundance (Fahay 1978). Annual trends in reported eel catches by individual states are composed of three basic groups: catches continuing to decline below the long-term (1950-1998) mean, e.g., New Hampshire, Rhode Island, Connecticut, New York, South Carolina; catches increasing or remaining above the long-term mean, e.g., Maryland, Delaware; and catches that are below the long-term mean but

2 have returned to values typical of those reported prior to the peak catches of the 1970s and early 1980s, e.g., Maine, Massachusetts, New Jersey, Virginia, North Carolina, Georgia, Florida (Jessop 1997) (Figure 2.1.2.7). Trends were assessed visually in relation to the long-term mean indicated for each state.

Caribbean and Central/South America: Little is known about the extent of yellow/silver eel fisheries and catches in Caribbean and Central and South American waters. Anecdotal reports indicate that occasional catches of some magnitude may occur in some areas. As noted in Section 2.1.1, the FAO reports of “river eel” catch are believed to be glass eel/ elver.

2.1.3 Reliability of reported catches

Canada: The elver fishery in the Scotia-Fundy area is closely regulated and of small size. Each fisher is required to provide daily catch and effort data. Independent dock-side monitoring was instituted in 1998. Illegal fishing at a low level has occasionally been reported and enforcement action taken. Consequently, misreported catches of elvers in Canada are believed minor.

The extent of the mis- or unreported catch of yellow/silver eels in Canadian provinces is unknown but often believed to be low. In Lake Ontario, the unreported eel catch is believed less than 5%. In Québec and the Scotia-Fundy portion of the Maritime Provinces, unreported catches are also believed low. For the St. Lawrence silver eel fishery, the extent of unreported eel catch was estimated at 8% in 1996 by comparing catches from index fishermen with official statistics. Sources of unreported catches might include recreational catches using a variety of gears, personal consumption, or local sale of commercially caught fish, native food fisheries, and incomplete data collection by statistical agencies. In the southern Gulf of St. Lawrence, a substantial portion of reported catch is subjectively estimated rather than tallied from purchase records, and substantial discrepencies between official figures and results of phone surveys have been reported. This suggests that reported landings for southern Gulf of St. Lawrence may contain substantial error.

United States: Mandatory reporting of catches in the Maine elver fishery was established only in 1998. Underreporting of catches occurred to an unknown, but possibly substantial, degree before 1998. The extent of underreporting of eel catches in American states is also unknown but is a major issue in some states, e.g., Maine, Maryland. Underreporting sometimes reaches 30% or more of the reported catch. Even when catch reporting is mandatory, misreporting may occur.

Caribbean and Central/ South America: The extent of eel fishing and reliability of catch data (presumably for elvers) in countries south of the United States is largely unknown.

2.2 Trends in Abundance Indices

2.2.1 Recruitment stages

The East River, Sheet Harbour (Nova Scotia) elver abundance series shows some variability over 10 years but no trend (Table 2.2.1.1; Figure 2.2.1.1). It represents the longest elver series available for this species. The East River, Chester (Nova Soctia) elver abundance index began in 1996 and is correlated with the East River, Sheet Habour. Index (n=4, r=0.998; p=0.002). Young yellow eels, recruiting to Lake Ontario at ages of more than three years, ascend the Moses- Saunders eel ladder on the upper St. Lawrence River. Eels have been counted at the ladder since 1974. Eel recruitment at the ladder has collapsed in recent years as indicated by the three order of magnitude decline in the number of recruits since 1986 (Table 2.2.1.2; Figure 2.2.1.2). Another abundance index series of recruiting young eels to the Petite Rivière de la Trinité, a small stream of the north shore of the Gulf of St. Lawrence (Québec), was carried out from 1982 to 1985, from 1993 to 1996, and reinstated in 1999 and 2000. These data show large inter-annual variability but no trend (Table 2.2.1.2; Figure 2.2.1.2). Finally, a short elver series was conducted in Little Sheepshead Creek (New Jersey) from 1989 to 1994, which suggests an upward trend in abundance in 1993 and 1994 (Table 2.2.1.2).

2.2.2 Yellow and silver eel stages

Three types of estimates of abundance trends of yellow and silver eels are available for the St. Lawrence River basin, one for waters of the State of New Hampshire, four for the Hudson River basin, and three for waters of the State of Virginia. Many of these time series were analyzed both with a nonparametric trend analysis, namely Kendall’s correlation tau as the test statistic on ranked data and by a resampling procedure. When available, probability values for the analysed data are given.

3 Lake Ontario trawl survey: A scientific trawl survey has been conducted in the Bay of Quinte on the north shore of Lake Ontario annually for the 28-year period from 1972 to 1999, except in 1989, for which year an estimate was used. The geometric mean number of yellow eels caught in ¼ nautical mile tows of a ¾-scale Yankee bottom trawl was calculated for each year (Table 2.2.2.1). The trend in this abundance index has been downward, although catches show considerable inter-annual variability (Figure 2.2.2.1). Catches in the last five years are an order of magnitude lower than in the first five years of the survey period.

Lake Ontario electrofishing survey: Records of catch per unit of effort were obtained for a commercial electrofishing operation for the 16-year period from 1984 to 1999. The mean number of eels caught per hour of electrofishing was calculated for each year (Table 2.2.2.1). Since 1989, the catch per unit of effort has declined significantly (p=0.0034) to less than 20% of the pre-1990 values (Figure 2.2.2.1).

Lower St. Lawrence weir silver eel survey: Large commercial weirs (interception nets) set perpendicular to the shoreline in the intertidal zone of the St. Lawrence estuary have been operated since the early 1900s. The fishery is prosecuted primarily in the 80 km extending downstream from about 120 km below Québec City. For the 21-year period from 1979 to 1999, the catch per unit of effort has been calculated as the kilograms of eels caught per meter of weir length (Table 2.2.2.1). The catch per unit of effort in the last five years has declined significantly (p=0.0015) to less than one-third of that of the first five years of the survey (Figure 2.2.2.1).

New Hampshire pot fishery survey: For 11 years in the period 1988 to 1999 (except 1989), catch per unit of effort was estimated for the commercial yellow-eel pot fishery in two to seven New Hampshire river basins (Table 2.2.2.1). The number off eels per pot hour (x10) increased from 1990 to 1995 and has decreased since (Figure 2.2.2.2). The shortness of the time series and the variability in fishing areas included in the survey makes interpretation problematic; there is no statistically significant trend.

Hudson River Utilities beach seine survey: For the 12-year period from 1985 to 1996, bi-weekly (June-October) scientific beach seine samples were taken in the brackish and tidal-freshwater portions of the Hudson River estuary. About 1,000 samples were taken each year with a 30.5 m beach seine. Annual catch per unit of effort was calculated as the total number of eels caught divided by the number of samples taken that year (Table 2.2.2.1). The catch per effort declined significantly over this period (p=0.0038) (Figure 2.2.2.3).

Hudson River Utilities fall shoal survey: For the same period and in the same area, bi-weekly samples were taken with a 1m2 Tucker trawl in the channel and a 3m beam trawl on the bottom in the channel and near shore. Annual catch per unit of effort was calculated as the number of eels caught divided by the volume of water sampled (Table 2.2.2.1). The catch per effort declined significantly over this period (p=0.0110) (Figure 2.2.2.3).

Hudson River Utilities impingement survey: At two power plants (Danskammer and Roseton) located near one another on the tidal freshwater portion of the Hudson River estuary, the total annual impingement of yellow and silver eels has been tabulated for the 23-year period from 1974 to 1996 (Table 2.2.2.1). Both series showed a decline over the period, with the downward trend more pronounced at Danskammer (Figure 2.2.2.3).

North Anna River electrofishing survey: Annual scientific electrofishing surveys were conducted in the Anna River, part of the York River basin, Virginia, for the 19-year period from 1981 to 1999. The catch per unit of effort is tabulated as mean number of eels caught per 70 m of electrofishing distance along the river (Table 2.2.2.1). This series shows an increasing but nonsignificant trend of abundance (Figure 2.2.2.4).

Potomac River commercial eel pot survey: For the 12-year period from 1988 to 1999, a catch per unit effort has been calculated from the commercial pot catches of yellow eels in the Potomac River estuary. The catch per unit of effort is tabulated as the number of eels caught divided by the number of pot-days fished (Table 2.2.2.1). There was a significant (p=0.0233) increase in catch per effort over this period (Figure 2.2.2.4).

Virginia Institute of Marine Science trawl survey: Annually during the 45-year period from 1955 to 1999, a scientific trawl survey has been conducted at several locations in the estuaries of the James, York, and Rappahannock Rivers of the Chesapeake Bay. A liner was added to the standard trawl in 1972, and a tickler chain was added in 1978. These changes undoubtedly altered the vulnerability of eels to the gear. Data are tabulated as the geometric mean number of eels >153 mm long per trawl (Table 2.2.2.1). Since the tickler chain was added, catch per effort increased for a few years, but it has declined since the mid-1980s (Figure 2.2.2.4).

Indices of abundance from a particular geographic area may not be independent of each other. Consequently, the common trend for each geographic area has been considered. In the upper St. Lawrence River and Lake Ontario,

4 juvenile eel indices declined, no trend was found in New Hampshire, the Hudson River indices declined (one non- significantly), the Virginia North American River index showed no trend, and the Chesapeake Bay trawl survey indicated a decline. Only the Potomac River index showed an increasing trend.

2.3 Causes of Observed Trends in Recruitment

In Section 2.2, the available trends in recruitment have been described. In summary, a major decline in recruitment has been observed in the Upper St Lawrence and Ontario region, but in other regions, declines have not been observed but data are limited and usually short term. The decline in the St Lawrence region is strongly paralleled by a steep decline observed in the recruitment of the European eel (Anguilla anguilla) in almost all of mainland Europe since 1980. A number of potential explanations have been suggested (Castonguay et al. 1994). These include natural (below) and anthropogenic, (Sections 3.2 and 3.3) as well as continental and oceanic causes. No conclusive evidence has been brought forward so far. The following sections describe the hypotheses raised.

2.3.1 Oceanic factors

The basin-scale water circulation in the North Atlantic Ocean is subject to decadal time-scale variation. The influence of large spatial and temporal scale variation and decadal trends in circulation in the western North Atlantic on survival of leptocephali and recruitment of glass eels has not been determined at this time. Nevertheless, the potential for such an influence is considered below.

General trends in circulation can be inferred from atmospheric patterns and from Gulf Stream patterns. The North Atlantic Oscillation Index (NAOI) is the sea level atmospheric pressure at the Azores or at Lisbon, Portugal, minus the sea level atmospheric pressure at Reykjavik, Iceland. A winter time series from 1864-1999 of the NAOI, normalised by dividing by the standard deviation of the values, is available (http:/www.cgd.ucar.edu/cas/climind/nao_winter.html). The cumulative sum of the index reveals several trends (Figure 2.3.1.1). From 1864 to about 1900, little trend occurred. Between 1900 and 1925, the NAOI was generally high, resulting in a rising trend in cumulative sum. The NAOI fluctuated about the average until about 1955, whereupon the NAOI was generally low until 1970 (Figure 2.3.1.2). The NAOI again became generally high between 1970 and 1995.

High values of the NAOI indicate periods of stronger winds, greater surface-water mixing, and a reduced and more easterly path of the Gulf Stream and North Atlantic Current, and vice versa. Changes in transport during decades of high or low values of the NAOI could influence the delivery of leptocephali of the American eel, especially at the northern end of its geographic range.

A five-year fast Fourier transform was fitted to the time series of the NAOI to filter out the high frequency variation and to emphasise the fundamental long period of approximately eight years in the atmospheric-ocean interaction. There is a strong negative correlation between the smoothed values of the NAOI and recruitment of glass eels of the European eel to den Oever, Netherlands (Figs. 2.3.1.3,4). Similarly, there is a negative correlation of the NAOI and the abundance of juvenile American eels ascending the Moses-Saunders dam between New York and Ontario, when the abundance data are lagged by four years (Figure 2.3.1.3).

The position of the North Wall of the Gulf Stream since 1966, derived from aircraft and satellite observations, has been published (http://www.pml.ac.uk/gulfstream/inetdat.htm). A useful index is based on principal components analysis of the latitude of the North Wall at six longitudes in the western North Atlantic (NWI) (Taylor 1995; 1996). A high NWI indicates relatively high latitude for the North Wall, and vice versa. The cumulative sum of the NWI shows that the trend between 1970 and 1982 was for southerly positions of the North Wall and between 1982 and 1995 was for northerly positions (Figure 2.3.1.5). The NAOI and the NWI are correlated with the latter lagging by about two years.

High values of the NWI imply stronger transport by the Gulf Stream in the western North Atlantic (Kelly 1991), but also stronger recirculation in the western North Atlantic (Kelly and Watts 1994). Among other influences, trends in the NWI are associated with trends in zooplankton abundance, at least in the eastern North Atlantic (Taylor 1995). If similar trends in production occur in the western North Atlantic, they could influence survival of leptocephali at sea. The NAO influences hydrographic conditions in the Sargasso Sea, which could influence the location and timing of spawning of eels and transport of the resulting leptocephali.

5 2.3.2 Effects on continental life stages

2.3.2.1 Continental factors affecting the incoming recruitment

Glass eels: The recruitment of young eels to the continent is monitored just inside continental waters. All fisheries for glass eel or elvers are operated at the interface between fresh water and the sea. Therefore, it seems rather unlikely that continental factors have been influencing the abundance of incoming young of the year already at sea.

Young yellow eels: Recruitment to the upper St. Lawrence and Lake Ontario as monitored at the Moses-Saunders dam involves young eels, usually ranging in age from 3-14 years, after their metamorphosis from the leptocephalus stage. Over the past 18 years, eel passage has decreased by three orders of magnitude. It is unclear to what extent this immigration has been affected by the Beauharnois Dam 50 km downstream, a reduction of vessel lockage (by half) in the St. Lawrence Seaway system during the period, and in increase in the commercial harvest (doubled) of yellow eels in Lake St. Francis, immediately downstream. These factors may have contributed to the decline but cannot fully explain this virtual loss of recruitment. Just as important for this more distant stock, oceanic effects (Figs. 2.3.1.3, 4) and changes in spawning stock abundance may be involved. However, since the St. Lawrence River is one of the largest rivers in North America, the decline in juvenile immigration to the upper St. Lawrence River-Lake Ontario area and the consequent reduction in yellow eel abundance and silver eel escapement has resulted in a major loss of escapement of large, fecund spawners from the continent.

2.3.2.2 Continental factors affecting the fecundity of the spawning stock

Continental factors affecting older continental life stages all operate through a reduction of the total fecundity of the escaping spawning stock, either in quality or in quantity of the escaping spawners.

Reductions in the quality of spawners escaping:

Parasitism- The swimbladder parasite Anguillicola crassus was found in the American eel (Barse and Secor 1999) in 1997, but may have been present earlier. This parasite is indigenous in eastern Asia, where it occurs in the Japanese eel (Anguilla japonica). It has been found in European eel in Europe since the mid 1980s. It has been hypothesised to hamper the oceanic migration of the silver eel to the spawning area. For the European eel, the occurrence of this parasite does not match the time that this decline in recruitment occurred. For the American eel, the apparently recentariul of Anguillicola crassus would not yet have had any substantial effect on the spawner escapement.

Contamination- Man-made organochlorines are bio-accumulated in eel, but there is no evidence of major effects on survival. Declines in recruitment in the St Lawrence River (and in Europe) do not coincide with periods of maximum contamination by organochlorine compounds (Castonguay et al. 1994; Knights 1996). Furthermore, spawner escapement from uncontaminated areas is supposed to be substantial.

Reductions in the number of spawners escaping: A number of hypotheses have been raised concerning a potential reduction in productivity of continental waters in terms of the number of spawners escaping. This might include reductions in habitat areas or in the accessibility of habitats, and intentional (fisheries) or unintentional (hydro-power dams; see Section 3.2) reductions of the continental stock by numbers.

Habitat area- Eels and eel fisheries are found all over the Atlantic region of North America, but most intensively in areas of maximum human environmental impact, namely the coastal zone of the continent. Land reclamation, flood control, and drainage of marshland all have reduced the amount of habitat available to the eel (Section 3.2.2).

2.4 Summary/Conclusions

American eels are exploited to varying degrees in one or more life stages (elver, yellow and silver eel) throughout much of their range. Fisheries for elvers are limited to localised areas of Atlantic Canada, the northeastern USA and several Caribbean countries. Elver catch series are short, show no trend, and are affected by market conditions. Yellow and silver eel catches from the Lake Ontario / St. Lawrence River ecosystem as well as from the Gulf of St. Lawrence have steadily declined by a factor of about three in the past decade. In contrast, catches from Atlantic Nova Scotia and the Bay of Fundy have increased by a factor of three since the mid or late 1980s. Yellow and silver eel catches in the USA exhibit various trends, primarily negative, but no geographic pattern occurs there as opposed to the patterns observed in Canada. Incomplete or misreporting is a major problem in the USA, but less so in Canada.

6 The two major abundance series of recruitment show dramatically different trends. A 10-year long abundance index series of elvers from Nova Scotia shows no trend. However, a longer 27-year series of recruiting yellow eels to Lake Ontario at an eel ladder shows a 3 order of magnitude decline that started in 1986. Eleven abundance indices from scientific surveys are available for yellow and silver eels. Eight of the 11 time series observed declines in abundance of yellow and silver eels. Only one series showed a significant increase with time. However, it was noted that indices of abundance from within a particular geographic area may not necessarily be independent of each other and that on the basis the six geographic units that were represented: 1) three indices declined (St. Lawrence/ Lake of Ontario, Hudson River and Chesapeake Bay); 2) two showed no trend (the New Hampshire and Virginia rivers), and 3) one increased (Potomac River).

A major decline in recruitment to the St. Lawrence / Lake Ontario has been observed that paralleled a steep decline in Anguilla anguilla in Europe. However, there is no evidence of a recruitment decline elsewhere due to lack of data for the American eel, although declines in abundance indices of yellow and silver eels were noted for 8 of the 11 time series in the species’ range. A number of potential explanations have been suggested for the St. Lawrence / Lake Ontario decline that includes anthropogenic continental causes as well as oceanic causes.

In conclusion, recruitment is down in the St. Lawrence / Lake Ontario, but the representativeness of this decline on the species as a whole is unknown. Further, the decline in yellow and silver eel abundance throughout the species’ range most likely represents a decline in escapement. Historically, the St. Lawrence River likely made an important contribution to the spawning population because it is one of the largest watersheds available to the species (Castonguay et al. 1994) and it produces large fecund females. A decline there results in a major loss of escapement of fecund spawners from the continent.

7 Table 2.1.1.1. Reported catch of elvers (kg) in the Bay of Fundy New Brunswick, Atlantic coast of Nova Scotia and in the State of Maine. Total State of Year NB NS Canada Maine 1977 10,000 1978 7,566

1989 0 26 26 1990 132 42 174 1991 65 0 65 1992 227 0 227 1993 534 179 713 1994 650 924 1,574 3,352 1995 549 2,689 3,238 7,545 1996 449 2,414 2,863 4,633 1997 852 3,270 4,122 3,345 1998 501 1,547 2,048 6,527 1999 0 478 478 1,630 2000 0 622 622 Note: little fishing activity occurred in 1999-2000 due to poor market conditions.

Table 2.1.1.2. Reported catches (t) of river eel in the Caribbean area, 1970-1997. Data are from FAO Fisheries Statistics. No catches were reported prior to 1975. It is believed that reported catches are Anquilla rostrata glass eel/elvers. Dominican Year Cuba Republic Mexico 1975 1 1976 7 1977 7 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1 1989 1 1 1990 1991 1992 1993 1994 49 1995 44 43 1996 35 1997 19

8 Table 2.1.2.1. Reported catches (t) of American eels in Canada, by province and region, exclusive of elvers. USL-LO- upper St. Lawrence R. and L. Ontario; SLR (Silver)- St. Lawrence River silver eels (inc. in Total Québec); NB/NS (Gulf)- Gulf of St. Lawrence area of New Brunswick/ Nova Scotia; NB/NS (Total) provincial total catch, nominally the sum of Gulf and SF area catches; Gulf/SF Total- sum of NB-/NS-Gulf and NB-/NS- SF catches. Total Canada = sum of USL-LO, Total Québec, NB(Gulf), NB(SF), PEI, NS(Gulf), NS(SF), and NFLD. USL- SLR Total NB NB NB NS NS NS Gulf SF Total Year LO (Silver) Québec (Gulf) (SF) (Tota PEI (Gulf) (SF) (Total) Total Total NFLD Canada l) 1884 2.52 2.5 1885 4.01 4.0 1886 8.73 8.7 1887 6.44 6.4 1888 4.14 4.1 1889 6.12 6.1 1890 8.42 8.4 1891 8.23 8.2 1892 19.62 19.6 1893 30.51 30.5 1894 35.15 35.2 1895 26.42 26.4 1896 32.31 32.3 1897 14.18 14.2 1898 18.72 18.7 1899 15.66 15.7 1900 18.18 18.2 1901 30.15 30.2 1902 29.39 29.4 1903 16.52 16.5 1904 20.52 20.5 1905 8.69 8.7 1906 8.28 8.3 1907 9.18 9.2 1908 10.08 10.1 1909 29.48 29.5 1910 47.25 47.3 1911 62.42 62.4 1912 101.43 101.4 1913 86.09 86.1 1914 134.96 135.0 1915 98.91 98.9 1916 64.26 64.3 1917 56.70 51.0 3.4 12.6 67.0 123.7 1918 61.38 61.8 23.9 13.1 98.8 160.2 1919 57.29 75.1 0.0 6.3 81.4 138.7 1920 41.36 56.1 282.3 24.2 8.2 10.7 43.1 366.8 1921 50.49 89.3 314.1 40.7 37.0 14.8 92.5 457.1 1922 66.11 20.7 454.5 14.0 9.5 7.5 31.0 551.6 1923 55.76 18.4 559.6 10.2 3.8 0.7 14.7 630.1 1924 53.91 50.9 540.6 10.0 35.6 7.5 53.1 647.6 1925 68.27 82.6 536.0 18.4 14.5 4.3 37.2 641.5 1926 51.8 113.8 960.1 5.4 8.7 5.7 19.8 1031.7 1927 48.33 72.1 615.5 1.4 5.9 3.5 10.8 674.6 1928 40.64 72.8 992.1 16.3 11.1 6.6 34.0 1066.7 1929 33.3 62.0 541.1 5.2 2.8 4.5 12.5 586.9

9 Table 2.1.2.1 cont’d USL- SLR Total NB NB NB NS NS NS Gulf SF Total Year LO (Silver) Québec (Gulf) (SF) (Total) PEI (Gulf) (SF) (Total) Total Total NFLD Canada 1930 43.43 67.3 596.7 11.8 5.9 15.2 32.9 673.0 1931 32.58 65.3 785.7 18.8 4.8 11.9 35.5 853.8 1932 22.28 80.8 864.7 9.1 8.2 13.8 31.1 918.1 1933 28.44 99.7 1115.0 11.0 5.9 16.7 33.6 1177.0 1934 22.19 82.7 1027.1 11.3 5.1 5.8 22.2 1071.5 1935 26.82 83.9 1026.9 7.6 10.2 5.2 23.0 1076.7 1936 23.67 84.0 988.2 4.4 8.6 4.1 17.1 1029.0 1937 29.7 80.1 836.9 5.7 16.5 4.2 26.4 893.0 1938 19.04 81.9 850.2 9.4 9.8 9.7 28.9 898.1 1939 10.22 74.9 774.0 11.0 9.6 8.7 29.3 813.5 1940 14.49 27.2 399.3 4.8 21.5 10.6 36.9 450.7 1941 7.38 21.8 453.7 3.6 15.7 3.4 22.7 483.8 1942 7.07 23.4 469.9 12.5 7.0 4.8 24.3 501.3 1943 15.62 29.7 638.5 14.1 10.9 5.0 30.0 684.1 1944 17.91 24.8 293.6 13.9 6.3 6.9 27.1 338.6 1945 19.58 2.0 389.6 14.5 6.1 10.7 31.3 440.5 1946 15.62 5.2 323.1 29.1 6.5 16.7 52.3 391.0 1947 15.03 3.0 324.4 31.8 35 7.7 13.6 34 53.1 392.5 1948 17.82 13.3 229.1 29.0 29 29.0 8.7 29 66.7 313.6 1949 20.7 41.4 185.8 29.4 32 15.9 37.6 87 82.9 289.4 1950 12.92 41.6 300.6 22.2 38 10.5 23.6 42 56.3 369.8 1951 21.42 54.7 351.4 15.5 15 13.6 20.9 46 50.0 422.8 1952 29.21 58.0 395.4 15.8 25 14.2 11.9 48 41.9 466.5 1953 25.61 63.0 406.1 13.1 34 33.1 7.7 48 53.9 485.6 1954 35.1 67.8 352.9 33.1 43 16.5 6.4 58 56.0 444.0 1955 30.6 68.6 405.8 48.6 88 24.1 10.5 78 83.2 519.6 1956 18.64 82.7 395.5 10.5 13 17.8 14.6 63 42.9 457.0 1957 44.6 94.9 562.3 8.6 14 12.3 10.1 32 31.0 637.9 1958 53.12 133.4 479.1 14.5 22 18.7 14.1 53 47.3 579.5 1959 55.39 102.7 390.3 23.6 30 26.4 11.4 31 61.4 507.1 1960 49.65 209.0 466.7 30.9 45 31.9 23.6 45 86.4 602.8 1961 58.63 167.0 388.7 57.4 66 17.7 27.8 52 102.9 550.2 1962 48.83 185.1 386.8 81.9 90 13.1 26.4 51 121.4 557.0 1963 76.32 197.1 472.8 53.7 66 15.9 23.6 68 93.2 642.3 1964 101.57 200.5 450.9 56.4 66 34.2 18.8 64 109.4 661.9 1965 84.92 257.8 553.8 62.6 68 48.6 16.3 56 127.5 766.2 1966 64.39 243.1 504.0 99.2 119 32.8 15.0 61 147.0 715.4 1967 61.04 250.8 447.0 108.0 98 61.8 52.3 67 222.1 730.1 1968 77.82 269.9 511.2 150.6 134 130.7 246.6 57 527.9 1116.9 1969 76.07 329.7 529.6 214.2 256 194.5 314.3 62 723.0 1328.7 1970 65.63 251.5 314.0 294.7 295 239.9 274.7 73 809.3 1188.9 1971 75.62 257.2 315.2 319.4 325 351.4 291.6 101 962.4 1353.2 1972 122.09 187.9 317.0 272.8 288 272.8 203.5 62 749.1 1188.2 1973 84.74 210.3 309.9 220.4 237 157.2 28.0 48 405.6 800.2 1974 99.86 247.5 395.5 156.2 178 101.2 265.5 42 522.9 1018.3 1975 166.29 368.3 529.4 120.8 169 103.5 374.1 41 598.4 1294.1 1976 153.96 274.8 425.3 118.7 198 94.1 279.0 27 491.8 1071.1 1977 186.39 362.7 510.9 110.1 265 97.6 5.9 15 213.6 910.9

10 Table 2.1.2.1 cont’d USL- SLR Total NB NB NB NS NS NS Gulf SF Total Year LO (Silver) Québec (Gulf) (SF) (Total) PEI (Gulf) (SF) (Total) Total Total NFLD Canada 1978 229.7 375.5 530.3 81.6 126 113.6 12.3 64 207.5 15.6 983.1 1979 221.53 344.3 510.4 102.4 223 111.0 12.6 32 226.0 23.4 981.3 1980 164.03 426.2 602.8 150.4 25 175 120.1 9.5 40 51 280.0 67 82.2 1196.0 1981 107.9 401.4 603.5 191.2 35 227 220.0 7.5 20 28 418.7 55 41.5 1226.6 1982 29.02 231.7 936.5 159.2 3 163 167.6 11.3 15 26 338.1 18 36.7 1358.3 1983 75.56 220.9 353.0 97.4 1 98 150.5 9.6 19 31 257.5 20 28.0 734.1 1984 122.4 281.6 413.7 122.4 3 125 164.6 8.9 8 17 295.9 11 14.0 857.0 1985 103.95 0.0 a 202.4 74 276 139.4 5.1 7 12 346.9 81 25.0 556.9 1986 116.1 320.0 504.4 230.2 54 284 226.0 15.6 6 20 471.8 60 26.6 1178.9 1987 102.9 280.9 428.1 171.6 49 220 149.9 13.2 15 28 334.7 64 30.6 960.3 1988 105.3 288.0 440.5 233.5 135 369 124.7 24.7 14 38 382.9 149 60.8 1138.5 1989 121.49 281.7 433.0 209.0 116 326 69.5 30.2 6 32 308.7 122 83.5 1068.7 1990 119.04 295.6 473.9 149.3 91 240 123.7 20.8 5 26 293.8 96 146.6 1129.3 1991 116.96 259.7 395.7 130.4 88 218 126.6 34.7 39 74 291.7 127 133.9 1065.3 1992 123.02 180.0 283.2 119.5 59 193 54.0 55.9 62 118 229.4 121 89.9 846.5 1993 104.96 203.8 330.5 88.2 116 204 73.9 89.1 71 160 251.2 187 116.1 989.8 1994 82.31 161.4 255.1 68.0 131 205 45.8 42.1 100 141 155.9 231 110.9 835.2 1995 62.17 142.7 278.0 60.0 114 140 34.3 16.2 118 123 110.5 232 85.4 768.1 1996 56.75 105.7 225.0 48.6 102 151 33.3 11.4 72 83 93.3 174 95.0 644.1 1997 43.38 99.9 202.0 34.6 111 147 44.0 17.1 64 81 95.7 175 72.0 588.1 1998 21.48 127.2 227.5 49.2 88 137 40.7 15.0 75 90 104.9 163 73.0 589.9 1999 20.52 93.1 167.0 38.4 79 117 52.0 9.0 67 78 99.4 146 55.0 487.9 aThe catch of silver eel was not recorded in 1985.

11 12

Table 2.1.2.2. Reported catches (t) of American eels, by state, for the Atlantic coast and Gulf of Mexico of the United States, 1950-1998. Data obtained from the National Marine Fisheries Service web site. Regional zones are: North Atlantic – ME-CT, Mid-Atlantic – NY-VA, South Atlantic – NC-FL, Gulf – LA-TX. North Atlantic Mid-Atlantic South Atlantic Gulf Year ME NH MA RI CT NY NJ DE MD VA NC SC GA FL.E FL.W FL.Tot LA TX Total 1950 17.2 7.9 15.1 6.8 123 34.6 15.4 422.8 227.5 79.2 4.4 0.6 5 954.5 1951 24.0 11.2 15.2 5.8 165.3 19.5 2.4 370.1 192.4 14.8 7.2 0.3 7.5 828.2 1952 19.1 11.9 6.9 8.3 70.1 22.4 15.5 381.6 167.1 17.5 5.5 5.5 725.9 1953 18.5 11.3 14.8 11.4 65 23.4 18.8 302.2 145 23.8 0.8 3.1 1.3 4.4 639.4 1954 5.2 9.8 6.9 11.4 81.1 30.8 16.4 165.7 190.5 16.8 0.8 2 3.9 5.9 541.3 1955 14.8 2.7 10.2 8.5 5.9 112.7 42.2 4.9 203 193 20.5 2.6 2.3 4.9 623.3 1956 13.3 2.3 7 9.4 8.5 106.9 26.7 6.8 211.6 203.3 51.3 0.2 2.3 2.4 4.7 652.0 1957 8.9 4.1 4.5 4.8 11.4 92.5 34.7 17.4 192.6 155.7 28.6 3 3.5 0.2 3.7 561.9 1958 9.7 3.4 8.3 9.4 9.4 122.3 28.3 11.7 182.2 190.3 39.4 0.1 3 0.2 3.3 617.8 1959 7.6 2.3 6 11.7 7.4 107.9 28.1 12 126.5 238.6 45 0.6 593.7 1960 13.8 2.7 16.2 19.6 8.5 103.4 18.8 2.9 88.5 87.2 29.9 0.4 391.9 1961 14.5 2.3 8.9 20.4 6 96.8 13.8 2.5 72.3 102 25.7 0.7 365.9 1962 16.1 2.9 7.3 15 8.1 58.7 10.1 4.3 51.8 96.3 19.6 0.1 290.3 1963 16.9 2.3 11 16.3 6.9 91.7 10.8 5.4 60.3 199.9 17.5 0.5 439.5 1964 10.4 2.3 8 14.2 8.2 78.8 35.7 5.4 84.4 142.3 24 2.5 55.4 55.4 471.6 1965 23.6 2.3 9.4 8.5 5.3 120.3 41.5 15.4 88.4 336.6 18.1 2.1 31.8 5.7 37.4 708.9 1966 22.1 3.2 11.2 11.3 3.9 77.2 65.2 14.5 100.8 212.7 24.7 18.7 0.4 13.8 13.8 579.7 1967 22.1 3.2 18.1 14.3 6.4 66.8 80.2 14.7 124.3 313.3 10.5 34.6 1.9 12.3 1.8 14.1 724.5 1968 29.7 16.4 20.5 17.6 19.1 65.7 53 16 120.2 321.9 11.1 55.2 0.7 21.6 0.4 22 769.1 1969 17.2 2.1 22.2 21 7.9 78.6 112.5 20 141.9 345.6 8 42.6 26.9 2.5 29.4 849.0 1970 17.1 2.5 25.6 16.6 22.5 62.7 95 26.4 131 546.5 7 10 2.4 13.6 13.6 978.9 1971 24.7 3.2 34.7 17.6 19.9 73.8 104.6 45.2 106.2 554.7 75.8 25.6 3 17.4 2.4 19.8 1109.0 1972 31.8 2.4 25.1 10.3 21.7 67.5 118.9 20.4 104.3 222.9 35.1 19.1 4.8 27.7 27.7 712.0 1973 34.5 2.5 17.9 8 12.1 51.6 104.7 27.4 81.8 115.5 60.6 31.9 5.1 37.1 37.1 590.7 1974 36.1 2.4 79.9 9.7 5.1 43.2 98.1 30.6 65.9 659.1 205 6.8 3.5 142.2 142.2 1388.0 1975 70.2 2.5 226.3 10 19.1 46.1 100.5 29.2 93 586.9 107.7 13.5 4.3 297.7 297.7 1607.0 1976 86.6 2.8 138.2 8.9 16.1 55.2 92.3 36.7 74.9 257.2 231.3 7.9 2 105.9 0.3 106.1 1116.0 1977 79.7 2.5 143.2 10.6 7.5 42.8 56.9 43.5 82.6 162.7 117.2 6.2 1.1 196.5 196.5 953.0 1978 60.6 2.0 150.4 11.9 11.5 48.2 53.9 85.3 93.5 527.7 315.5 1.1 241.7 241.7 1603.0 1979 50.4 1.9 135.8 10.2 12.3 44 52.8 85.3 121.7 544.6 433 4.5 3.9 144.7 144.7 1645.0 1980 47.9 2.7 114.7 10 10.9 209 38.1 60.3 146.2 193.2 435.5 45.1 51.7 2.9 54.7 1368.0 1981 25.0 3.0 97 8.2 10 154.9 53.1 90.4 330 342.1 197.8 5.8 51.6 0.5 52.1 1369.0 1982 20.5 3.2 60.1 4.9 8.8 83.2 109.9 61.9 38.3 323.9 215.7 16.6 30.4 30.4 3.5 980.9 1983 5.4 1.7 27.7 4.8 0.9 26.4 136 47.7 41.7 313.3 183.3 0.1 0.9 1 9 798.9 1984 1.3 14.9 2.8 3.4 48.2 242 56.4 49.7 356.7 320.4 0.7 0.7 2.7 0.1 1099.0 1985 10.9 1.0 11.6 3.9 2.1 53.7 153 60.3 39 359 101.7 1.2 1.2 11.5 808.9

12 Table 2.1.2.2 cont’d North Atlantic Mid-Atlantic South Atlantic Gulf Year ME NH MA RI CT NY NJ DE MD VA NC SC GA FL.E FL.W FL.Tot LA TX Total 1986 7.6 0.4 11.5 1.2 5.2 63.8 133.5 59.1 50.6 333.6 153.5 1.7 4.2 5.8 0.04 825.8 1987 0.2 11.2 0.5 11.2 28.8 86.8 43.3 56.3 338.5 58 0.02 634.8 1988 0.1 11.7 1.8 19.1 14.6 87 42.6 63.5 299.4 26 0.7 0.4 5.5 5.9 1.1 0.01 573.5 1989 12.7 13.6 0.9 9.6 20.2 87.9 52.5 149.2 237.2 69.2 2.5 0.1 3.5 3.6 0.4 659.5 1990 30.0 12.6 6 18.9 61.1 117.5 125 124.7 25.6 0.5 1.6 2.1 523.5 1991 8.3 10.6 4.4 20.2 98.4 116.2 176 219.1 5.5 0.7 0.2 2.8 3.1 0.02 662.5 1992 8.0 0.1 16.2 6.3 26.8 84.8 31.7 122 280.4 8 0.4 0.5 0.9 585.2 1993 6.6 0.6 12.6 4.5 19.3 88.2 59.9 180.4 274.1 14.8 0.8 0.6 1.4 662.4 1994 24.1 6.9 76.2 97.5 237.3 193.9 43.5 0.5 0.1 0.6 680. 1995 23.7 0.01 5.8 0.2 66.1 73.1 180.4 144.2 77 0.2 570.7 1996 4.6 0.01 0.1 8.3 0.4 105.1 186.7 276.3 64.2 645.7 1997 21.1 0.1 0 52.9 187.3 94.4 58.4 414.2 1998 15.9 0.2 2.5 0.2 42.8 59.6 212.1 86.5 41.3 461.1 13 13 Table 2.2.1.1 Annual estimates, with 95% confidence intervals (CI), of the run of American eel elvers to the East River, Sheet Harbour, and the East River, Chester, Nova Scotia. Site Year Estimate 95% CI Site Estimate 95% CI East River 1989 10,700a Sheet Harbour 1990 218,300 1991 376,000 1992 219,200 1993 134,100 1994 309,900 ± 10,900 1995 101,500 ± 1,600 1996 336,500 ± 10,100 East River 1,138,100 ± 24,200 1997 467,400 ± 7,000 Chester 1,419,000 ± 52,100 1998 109,200 ± 2,000 432,400 ± 8,200 1999 134,600 ± 600 441,700 ± 9,800 aThe run size was greatly underestimated due to operational problems.

14 Table 2.2.1.2. Recruitment indices for American eels. Eel ladder index is number of eels (ages 3-14) passed per day during the 31-day peak period at the ladder at the Moses-Saunders Dam. Juvenile eel counts (yearlings and 2-year olds) at Petite Rivière de la Trinité, mid-June to August, north shore, St. Lawrence River. Elver counts East River are total numbers (x10,000) caught. Glass eel counts on Little Sheepshead Creek are numbers dip-netted per m3. Virginia index (VIMS) is catch per trawl with some modifications, including a liner added in 1972 and a chain tickler in 1978. St. Lawrence River system Nova Scotia New Jersey Virginia Eel ladder Petite rivière East River Little Virginia Inst. Upper St. Lawrence de la Trinité Sheet Harbour Sheepshead Cr. Marine Sci. Year (juveniles) (juveniles) (elvers) (glass) (<153 mm) 1955 0.414 1956 0.047 1957 0.000 1958 0.000 1959 0.000 1960 0.000 1961 0.000 1962 0.000 1963 0.000 1964 0.110 1965 0.072 1966 0.008 1967 0.052 1968 0.114 1969 0.504 1970 0.015 1971 0.048 1972 0.017 1973 0.049 1974 7934 0.039 1975 14403 0.082 1976 10363 0.000 1977 20013 0.045 1978 16448 0.023 1979 18977 0.652 1980 9046 0.571 1981 13796 0.205 1982 27489 4576 0.133 1983 26426 4389 0.748 1984 15051 1046 0.764 1985 18510 1117 0.070 1986 5380 0.324 1987 9276 0.634 1988 5442 0.000 1989 5795 18.1 0.251 1990 3096 21.8 20.8 0.608 1991 1226 37.6 17.2 0.261 1992 277 21.9 17.2 0.752 1993 232 3681 13.4 23.6 0.318 1994 4998 763 31 32.4 0.081 1995 671 4047 10.2 0.005 1996 405 5547 33.7 0.014 1997 144 46.7 0.082 1998 57 10.9 0.020 1999 27 933 13.4 0.002

15 Table 2.2.2.1. Indices of abundance of yellow and silver American eels. All indices refer to yellow eels except for the impingement ones, which are yellow and silver combined, and the lower St. Lawrence River weir, which is exclusively silver eels. The Lake Ontario trawl index is number per ¼-nautical-mile, Bay of Quinte, and the Lake Ontario electrofishing index is number per electrofishing hour in the vicinity of Main Duck Island. Lower St. Lawrence River eel catch is kg per m of leader. New Hampshire eel pot index is lb(x10) per pot hour. The beach seine survey is number of eels caught per year divided by number of sampling periods. The fall shoal survey is the number of eels caught per trawl divided by the volume of water sampled. Impingement indices are total annual catch. The North Anna River index is the number of eels per 70 m of electrofishing (updated data file NA00x56B courtesy of N, Wooding, Virginia Power, Richmond, VA). The Potomac River eel pot is the summed annual catch divided by pot days. The Virginia index (VIMS) is mean catch per trawl haul.

New York, Hudson River St. Lawrence River system Virginia New Impingement Lake L. Ont. Lower St. Hampshire Beach Fall N. Anna R. Potomac R. VA Inst. Ontario electro- Lawrence commercial seine shoal Dans- electro- commercial Mar. Sci. Year trawls fishing weir eel pot survey survey Roseton kammer fishing eel pot (>153 mm)

1955 6.55 1956 1.95 1957 0.16 1958 2.12 1959 0.60 1960 0.18 1961 1.88 1962 1.21 1963 0.80 1964 0.82 1965 0.26 1966 0.39 1967 0.77 1968 1.46 1969 1.04 1970 0.21 1971 1.23

16 Table 2.2.2.1 cont’d

St. Lawrence River system New York, Hudson River Virginia Lake L. Ont. Lower St. Hampshire BeachFall Impingement N. Anna R. Potomac R. VA Inst. Ontario electro- Lawrence commercial seine shoal Dans- electro- commercial Mar. Sci. Year trawls fishing weir eel pot survey survey Roseton kammer fishing eel pot (>153 mm) 1972 1.873 0.63 1973 1.620 4.2 0.96 1974 0.997 20.7 119.6 0.40 1975 1.543 135.9 94.4 0.87 1976 1.286 92.6 231.9 0.53 1977 1.064 151.8 253.7 1.13 1978 0.417 63.3 122.5 0.98 1979 0.767 8.25 63.5 103.1 1.98 1980 0.252 10.08 64.0 186.2 4.19 1981 1.530 10.02 62.8 144.6 6.50 2.30 1982 1.884 18.15 23.1 173.1 6.25 2.36 1983 0.557 12.14 42.7 219.8 4.50 9.29 1984 0.330 85.6 7.66 17.2 150.2 7.00 5.66 1985 0.778 63.1 7.42 0.32 1.04 16.9 139.3 7.25 8.11 1986 0.865 82.9 6.79 0.16 1.39 11.8 107.2 4.00 4.56 1987 1.552 89.0 5.61 0.11 1.15 37.1 189.5 9.75 3.14 1988 0.299 68.8 5.52 0.58 0.14 1.24 40.6 110.9 4.00 1.13 1.32 1989 0.952 93.0 5.62 0.10 0.86 19.8 84.2 10.75 1.14 7.47 1990 0.356 64.1 7.05 0.12 0.08 0.20 41.1 91.6 9.00 1.05 4.51 1991 0.454 38.5 5.84 0.24 0.21 0.59 50.4 87.4 7.00 0.98 2.01 1992 0.584 44.5 3.78 0.39 0.01 0.01 36.4 65.3 5.00 1.62 7.01 1993 0.434 22.7 4.36 0.38 0.10 0.84 19.7 59.8 4.00 1.43 2.96 1994 1.157 30.0 3.57 0.51 0.09 0.98 22.6 46.2 2.75 1.05 2.14 1995 0.091 10.5 3.14 1.21 0.04 0.12 94.9 95.0 7.25 1.69 2.32 1996 0.356 14.9 3.58 0.59 0.01 0.01 27.3 62.1 15.50 1.72 2.55 1997 0.085 7.3 3.22 0.34 11.00 1.71 2.17 1998 0.123 12.9 4.25 0.10 10.25 1.76 1.97 1999 0.074 21.6 4.02 0.06 11.75 1.52 1.30

17 3500

Canada USA Total 3000

2500

2000

Catch (t) 1500

1000

500

0 1950 1956 1962 1968 1974 1980 1986 1992 1998

Figure 2.1.2.1. Reported catches of American eel (yellow and silver) by Canada and the United States, with a combined North American total, 1950-1998.

18 1500

Canada ON/PQ Gulf SL 1250

1000

750 Catches (t)

500

250

0 1875 1900 1925 1950 1975 2000

Figure2.1.2.2 Reported catches (t) of yellow and silver American eels from southern Gulf of St. Lawrence, (Gulf SL: Prince Edward Island, and portions of New Brunswick, and Nova Scotia), from Ontario and Quebec (ON/PQ) and from all of Canada, 1884-1999 (data in table 2.1.2.1).

19 1000 USL-LO 800 SLR (silver) Quebec

600

400 Catch (t) Catch 200

0

1950 1960 1970 1980 1990 2000 Year

Figure 2.1.2.3 Reported catches of American eels in the upper St. Lawrence River-Lake Ontario (USL-LO), silver eels in Québec and total yellow/silver eels in Québec, 1950-1999.

20 1 1970 - fishery closed east of long. 76°50' due to mercury contamination 2 1972 - harvest increased because export markets approved 3 1982 - closure of European market due to contaminants 200 4 1985 to 1988 - commercial licence buyout by OMNR 5 1998 to 1999 - reduced effort, size limits and some zone closures due to contaminants

2

100 4 Commercial catch ('000 kg)

1

3 5

0 1880 1900 1920 1940 1960 1980 2000

Figure 2.1.2.4. Commercial catch of American eels for all statistical districts in the Ontario waters of Lake Ontario and the upper Saint Lawrence River, 1884-1999. Major events that could affect the fishery and catch are indicated.

21 NB 400 PEI NS 350 NFLD 300

250

200

150 Catch (t) 100

50

0

1950 1960 1970 1980 1990 2000 Year

Figure 2.1.2.5. Reported catches of American eels, by province, in Atlantic Canada, 1950-1999. The silver eel catch was not recorded in 1985.

22 250 250 Gulf N.S. Scotia-Fundy N.S. 200 200 Golfe N.E. Scotia-Fundy N.E. 150 150 100 100 50 50

Catch/Prises (t) Catch/Prises 0 0 1980 1990 2000 1980 1990 2000

) 250 250 t

( Scotia-Fundy N.B. Gulf N.B.

s 200 200

e Golfe N.B. s

i 150 150

r P

/ 100 100

h c

t 50 50 a

C 0 0 1980 1990 2000 1980 1990 2000

250 Gulf P.E.I. 200 Golfe I.P.E. 150 100 50

Catch/Prises (t) 0 1980 1990 2000

Figure 2.1.2.6. Annual catches (t) of American eels in the Gulf and Scotia-Fundy areas of the Maritime Provinces, 1980-1999.

23 MAINE 120 DELAWARE 80 80 40 40 0 0 20 NEW HAMPSHIRE MARYLAND 400

10 200

0 0 MASSACHUSETTS VIRGINIA 200 600 400 100 200 0 0

) NORTH CAROLINA t RHODE IS. ( 400

20 H

C 10 200

T A

C 0 0 CONNECTICUT 60 SOUTH CAROLINA 20 40 10 20 0 0

NEW YORK GEORGIA 200 40 100 20 0 0

NEW JERSEY 300 FLORIDA 200 200 100 100 0 0 1950 1960 1970 1980 1990 2000 1950 1960 1970 1980 1990 2000

Figure 2.1.2.7. Annual reported catches of American eels, by state, for the Atlantic coast, 1950-1998. The horizontal line in each panel is the long-term mean catch for that state.

24 1000

North Atlantic Region ME 800 NH MA

600 RI CT

400 Catch (ton)

200

0 1950 1960 1970 1980 1990 Year

Figure 2.1.2.8. Reported catches of American eels in the North Atlantic region of the Atlantic coast of the United States.

NY Mid Atlantic Region 1000 NJ DE MD 800 VA

600

400 Catch (ton)

200

0 1950 1960 1970 1980 1990 Year

Figure 2.1.2.9. Reported catches of American eels in the Mid Atlantic region of the Atlantic coast of the United States.

25 1000 South Atlantic Region NC 800 SC GA FL 600

400 Catch (ton)

200

0 1950 1960 1970 1980 1990 Year

Figure 2.1.2.10. Reported catches of American eels in the South Atlantic region of the Atlantic coast of the United States.

100

Gulf Region LA 80 TX

60

40 Catch (ton) Catch

20

0 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 Year

Figure 2.1.2.11. Reported catches of American eels in the Gulf of Mexico region of the Atlantic coast of the United States.

26 1.5 Chester

1 Run size ('000 000)

0.5 Sheet Harbour

0 1988 1990 1992 1994 1996 1998 2000

Figure 2.2.1.1. Annual run size of American eel elvers to the East River, Sheet Harbour, and the East River, Chester, Nova Scotia, 1990-1999.

27 1970 1980 1990 2000 30000 6000

St Lawrence 25000 Trinité 5000

20000 4000

15000 3000

10000 2000 St Lawrence Petite rivière de la Trinité 5000 1000

0 0 1970 1980 1990 2000 Year

Figure 2.2.1.2. Annual indices of abundance of juvenile American eels from sites in the upper (Moses-Saunders eel ladder) and northern (Petite Rivière de la Trinité) parts of the Gulf of St. Lawrence.

28 1970 1980 1990 2000 2.5 100 20

St Lawrence 2.0 River system 80 15

1.5 60

10

1.0 40 Weir

Trawl CPUE Trawl trawl Electrofishing electro 5 0.5 20 weir

0.0 0 0 1970 1975 1980 1985 1990 1995 2000 Year

Figure 2.2.2.1. Annual indices of the abundance of American eels (yellow eels in the Lake Ontario trawl and Lake Ontario electrofishing surveys and silver eels in the St. Lawrence River weir survey) in the St. Lawrence river system.

1.5

New Hampshire

1.0

0.5 eel poteel CPUE

0.0 1970 1980 1990 2000 Year

Figure 2.2.2.2. Index of yellow eel abundance (CPUE x 10 lb/h) from New Hampshire eel pot fishery.

29 1970 1980 1990 2000 0.5 1.6 fall shoal Hudson 250 beach 1.4 River 0.4 Roseton 1.2 200 Danskammer 1.0 0.3 150

0.8

0.2 100 0.6 Beach seine Fall survey Impingement 0.4 50 0.1 0.2 0

0.0 0.0 1970 1975 1980 1985 1990 1995 2000 Year Figure 2.2.2.3. Indices of yellow eel abundance (eel impinged) at power station cooling water intakes at two Hudson River sites and of fall shoal and beach seine surveys in the lower Hudson River.

1970 1980 1990 2000 20 15

1.8 N Anna Virginia

15 Potomac 1.6 VIMS 10

10 1.4

VIMS Potomac

N AnnaRiver 5 1.2 5

1.0

0 0 1970 1980 1990 2000 Year

Figure 2.2.2.4. Annual indices of yellow eel abundance caught by trawl survey in Chesapeake Bay, Virginia, by electrofishing survey in the North Anna River, Virginia, and by commercial eel pot survey in the Potomac River, Virginia/Maryland.

30 25

20

15

10

5 Cumulative Winter NAOI Winter Cumulative 0

-5

-10 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 Year

Figure 2.3.1.1. Cumulative sum of the normalized December-March index of the North Atlantic Oscillation from 1864 to 1999.

10

5

0

-5

-10

-15 Cumulative Winter NAOI Winter Cumulative

-20

-25 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 Year

Figure 2.3.1.2. Cumulative sum of the normalised December-March index of the North Atlantic Oscillation from 1951 to 1999.

31 1970 1975 1980 1985 1990 1995 2000 4 -10 NAO 5 y FFTaverage NAO cum den Oever -8 -1.0 3 St Lawrence lag 4 year

-6 -0.5 2

-4 NAO 0.0 recruitment index recruitment 1 -2

0.5

0 0 1970 1975 1980 1985 1990 1995 2000 year

Figure 2.3.1.3. Recruitment indexes for glass eels of the European eel at den Oever, Netherlands, and for juvenile American eels at the Moses-Saunders dam on the St. Lawrence River, and the smoothed averages and cumulative sums of the North Atlantic Oscillation Index, 1970-1998.

1930 1940 1950 1960 1970 1980 1990 2000

2

-2 1

0 0 index -1 NAO 2

-2 den Oever index

NAO 5 y FFTaverage 4 -3

1930 1940 1950 1960 1970 1980 1990 2000 year

Figure 2.3.1.4. Recruitment index for glass eels at den Oever, Netherlands in Comparison with the smoothed average of the North Atlantic Oscillation Index, 1937-1998.

32 5 Annual index Cumulative sum

0

-5 North Wall of Gulf Stream Index Stream Gulf of Wall North

-10 1965 1970 1975 1980 1985 1990 1995 2000

Year

Figure 2.3.1.5. Annual index and cumulative sum of the index of position of the North Wall of the Gulf Stream in the western North Atlantic Ocean.

33 3 IMPACT OF FISHERIES IN SELECTED SYSTEMS ON EGG PRODUCTION

3.1 Impacts of Fisheries on Escapement

The American eel is subject to commercial fisheries at the glass eel, elver, yellow, and silver stages. This section presents the sparse information available on mortality from these fisheries, and its effects on subsequent egg production.

3.1.1 Elver fisheries

East River, Chester: The only estimates of exploitation rate of elver fisheries in North America were made in 1996, 1997 and 1998 at East River, Chester, Nova Scotia (watershed area 134 km2). Commercial dipnetters caught 350,500 elvers in 1996 455,500 elvers in 1997 and 224,200 elvers in 1998 (Jessop in press). Research traps located at a barrier to elver passage, located approximately 100 m upstream from the commercial fishery, took 787,600 and 963,500 and 8,200 elvers, respectively, in the three years. Exploitation rates (commercial catch/(commercial + research catch) were 30.8% in 1996, 32.1% in 1997 and 5.8% in 1998. (Table 3.1.1.1).

If there is no density-dependent change in sex ratio, growth, survival, or emigration rate in subsequent stages, the reduction in egg production due to the elver fishery will be equivalent to the percent elver exploitation. However, such density-dependent effect are believed to occur.

3.1.2 Yellow/silver eel fisheries

Maryland: In Maryland waters of Chesapeake Bay, eels collected from the fishery during spring 1999 were subject to an overall disappearance rate of 0.5, based on a catch curve derived from an age-length key created from aged fish. Disappearance includes natural mortality, fishing mortality, and emigration. Natural mortality was estimated at 0.25, based on Anthony's (1982) method (3/maximum age in the sample). This means that the sum of fishing mortality and emigration was 0.25.

Prince Edward Island: Fishing mortalities in exploited waters of PEI were estimated through comparisons of length and age structure in fished and unfished areas. Eels of lengths 50-80 cm were strongly represented in samples from unfished freshwater ponds, and in samples collected in 2000 in the Pinette River estuary which is currently unfished (Figure 3.1.2.1). Lengths >50 cm were poorly represented in samples from exploited tidal waters, and from a sample collected in the Pinette estuary in 1973, at a time when the area was subject to commercial fishing. These size ranges are presumed to be females only, based on sex ratios of eels of this size measured elsewhere in eastern Canada.

For each of these sites, lengths were converted to age using a von Bertalanffy equation derived from otolith aging of 83 eels on the Boughton River, PEI. Disappearance was calculated by simulating a population equivalent to the number appearing in the modal age (Figure 3.1.2.2). An instantaneous annual disappearance rate was then applied to this population, leading to a simulated series of numbers at age. Differences between this series and the actual age distribution were squared and summed. The annual disappearance rate for the simulated series was then adjusted, and the value that gave the lowest sum of squared differences was taken as the best estimate of disappearance rate.

Mean disappearance rates of exploited and unexploited sites were 0.76 and 0.26, respectively. Fishing mortality was estimated as the difference between disappearance rates in exploited and unexploited areas. Three comparisons were made: tidal exploited vs ponds, Pinette 1973 vs. Pinette 2000, and tidal exploited vs. Pinette 2000. Estimated fishing mortalities were 0.54, 0.45, and 0.51, respectively (Table 3.1.2.1). The mean was 0.50. The mean disappearance rate for unexploited sites was 0.26. This is equivalent to the sum of natural mortality and emigration.

The effect of the PEI eel fishery on spawning escapement was estimated through a life table model which tracks mortality, numbers, harvest, and emigration during each year of continental life, and calculates egg production from escaped females which reach the Sargasso Sea. Major assumptions and inputs given in Table 3.1.2.2.

Under these assumptions, exploitation reduced female escapement by 89.9%, and egg production by 91.9%. The greater percentage reduction in egg production was due to the cumulative effects of fishing mortality, which reduce large females (which lay more eggs) proportionately more than smaller females.

The estimated reductions in female escapement and egg production for PEI as a whole would be less than these values, because there is no eel fishery in non-tidal waters, and there is minimal fishing effort in the central and western portions of the Northumberland Strait, which amount to about one third of PEI's coastline.

34 The analysis presented above is subject to several sources of uncertainty. A major point of uncertainty stems from the calculation of age distribution from lengths, via the von Bertalanffy curve. There are three main problems.

1) Eel growth is highly variable, so individuals may deviate widely from the expected growth trajectory. 2) Eels available for aging are those that have not been fished and have not emigrated. Fishing mortality and emigration rate are both size-related, so there is a truncation of larger individuals. This means that age-length curves are largely based on an unrepresentative sample of slow-growing animals. 3) Eel growth shows high spatial variability. Hence the age-length relation used in this analysis, derived from a single system, may not be representative of other areas of PEI.

A second source of uncertainty is the assumption, implicit in the calculation of disappearance rate, that recruitment was uniform during the years that fish present in the samples arrived in the system. In fact, recruitment is unlikely to be uniform in any fish stock. This uncertainty may be reduced to some degree in comparisons of samples taken in the same years.

A third source of uncertainty is the assumption that growth, mortality, and emigration schedules are similar between the compared sites. This assumption is questionable for habitats which differ (i.e., tidal estuaries vs. freshwater ponds). It is more tenable for similar habitats (Pinette estuary vs. other bays and estuaries).

In addition, natural mortality and emigration rates are poorly known, and their uncertainties will decrease the reliability of estimated fishing mortality and loss of egg production.

Given the above uncertainties, fishing mortality and reduction in egg production as estimated for PEI eel fisheries must be considered as crude and preliminary. Nevertheless, the similarity of fishing mortality estimates from the various comparisons (range 0.45 - 0.54) supports the belief that fishing mortality in exploited PEI waters is substantial, and, due to the cumulative effects of multi-year exposure to fishing, leads to a major reduction in egg production.

3.1.3 Silver eels

St. Lawrence Estuary: The estuary of the St. Lawrence River, site of a major silver eel fishery, is the only location in North America where fishing mortality of migrating silver eels has been measured.

In 1996 and 1997, 1,047 and 1,433 silver eels were captured and marked while migrating from the freshwater St. Lawrence watershed. Downriver, 28,714 and 51,558 eels were examined for marks in each of the respective years. Analysis conducted with pooled data (Petersen estimate) and with stratified data (Schaefer and Darrock-Plante estimates) indicated migrating populations of 470,000 silver eels in 1996 and 380,000 in 1997.

Based on landings of 119.5 t in 1996 and 111.8 t in 1997, exploitation rates in this fishery were 25% (18-33%) in 1996 and 20% (14-28%) in 1997 (confidence ranges are α = 0.05). Escapement in the fishery was greater in 1996 (about 378,000) than in 1997 (about 288,000).

Because migrating silver eels are unlikely to face density-dependent effects in the ocean, it is probable that the 20-25% exploitation rates measured in this study produced similar decreases in egg production at the spawning grounds. It must be noted that exploitation at the migrating silver stage is only part of the anthropogenic constraints to which eels in the St. Lawrence system are subject. Prior to the estuarine fishery, these eels also face substantial mortality and habitat limitations due to upstream fisheries, dams, and pollution.

3.1.4 Natural mortality

Natural mortality estimates are required to evaluate the effects of fisheries on egg production because natural mortality rates influence life expectancy, and therefore the expectancy of spawning. Natural mortality has been estimated for elvers at one site, and for yellow/silver eels at two sites in North America.

East River, Chester: Jessop (2000) used trap counts and mark-recapture techniques to estimate that 99.45-99.68% of juvenile eels died between their arrival in May as elvers in the East River, Chester, Nova Scotia, and the subsequent October. This corresponds to instantaneous daily mortalities of 0.0612 and 0.0675. However, Jessop (2000) considered these mortalities to be abnormally high because of low pH in the river and heavy intra-specific predation.

35 Prince Edward Island: The analysis of length-frequency data for unexploited PEI waters gives estimates of 0.26 for natural mortality + emigration (Section 3.1.2 above), but this method does not separate the contributions of natural mortality and emigration.

Petite Trinité: Estimates of juvenile recruitment to and silver eel emigration from the Petite rivière de la Trinité, Québec, offer the opportunity to estimate cumulative natural mortality of the freshwater stage of the American eel. This river drains a watershed of 250 km2 into the northwestern Gulf of St. Lawrence. No fishery is permitted on the river.

Dutil et al. (1989) provided an index of recruitment to the river, consisting of three counts of juvenile eels per night, every fourth night, made at a rocky ledge adjacent to a waterfall near the river mouth (Table 3.1.4.1). The eels were dominated by length classes under 110 mm, representing primarily age 1+ (arriving elvers are considered age 0+). This series was resumed with the same methodology in 1993-1996. In 1999, the method was repeated, and the total run was estimated by mark and recapture methods.

Total runs for 1982-1985 were estimated by applying the ratio of the night counts in these years to the night counts in 1999, to the 1999 run estimate. This was done in two ways: by using the total of counts conducted every four days, and the mean of the highest 10 counts for the season. Total runs for 1982-1985 ranged from 14,014 to 61,308 (total count method) and 15,382 to 67,294 (best 10 method) (Table 3.1.4.1). Mean estimated runs were 37,272 and 40,912 eels by the total and highest 10 methods, respectively.

The number of silver eels exiting the river in 1999 was estimated at 800, based on the number caught when the river was partially blocked. These eels ranged from age 15 to 24 years; hence the silver eels emigrating in 1999 entered the river between 1976 and 1985.

If we assume that the estimates of river recruitment for 1982-1985 are also valid for the 1976-1985 recruitment years, and that the estimates of silver eel emigration for 1999 are also valid for other years in which the 1976-1985 recruits emigrated, we can calculate the mean in-river survival rate of these animals as 800/37,272 (2.0%, total method) and 800/40,912 (2.1%, best 10 method).

Natural mortality in fishes tends to decrease with body weight. Natural mortality as estimated by Lorenzen's (1996) general equation for fish produced very low cumulative survivorships (under 0.002%) for the period between age 1 and age 5. For all practical purposes, cohorts were extinct by age 10. To derive a weight-based function that was consistent with a cumulative 2% survival between age 1 and age 20, as estimated for Petite Trinité, the Lorenzen mortality was multiplied by an adjustment factor. The 2% cumulative survivorship was achieved when the adjustment factor was set at 0.164. Under this scenario, instantaneous natural mortality at age 10 was 0.13, which is far lower than the M of 0.79 for age 10 eels using the unadjusted Lorenzen function (Figure 3.1.4.1).

Chesapeake Bay: Natural mortality of Chesapeake Bay American eels during spring 1999 as estimated by Anthony's (1982) method (3/maximum age in the sample) was 0.25.

Comments: The natural mortality + emigration estimate for PEI (0.26) and the natural mortality estimate for Petite Trinité (0.13 at age 10) suggest weight-specific natural mortalities that are substantially lower than those of most fish species. The case for low natural mortalities may be strongest in eastern Québec rivers and some Atlantic Canada rivers such as the Lattave River (Jessop 1987) and upper Saint John River where eels may be 30-40 years old before migration. In which there is a long residency period for yellow eels, such as Petite Trinité.

The Petite Trinité natural mortality derived from the Anthony (1982) method was 0.125, which is similar to the value based on estimates of incoming and exiting eels (0.13 for age 10). The Anthony (1982) and Lorenzen (1996) methods are based on relations that apply to general fish populations, while the PEI and Petite Trinité estimates are based on specific eel data. This could account for the difference in estimates, as could differences between Canadian and Mid- Atlantic environments. If yellow eels have low natural mortality, management regimes that tie maximum permissible fishing mortality to natural mortality may be constrained to low exploitation rates.

3.2 Fish Passage

3.2.1 Potential effects of dams on migration

In order to complete their life cycle, eels require access to and from the sea. Therefore, dams have the potential to seriously affect eel populations in upstream freshwater habitats. Greeley (1932) hypothesised that an increase in dam construction in New York state streams could reduce eel abundance. One of the clearest examples of dam effect on the

36 Atlantic coast is on the Richelieu River where recruitment has been seriously impeded since the reconstruction of two dams in the 1960s (R. Verdon, pers. comm.).

Based on their extensive review of the literature, Richkus and Whalen (1999) concluded that dams also have indirect effects including on habitat, population density, growth rate and sex ratio clearest. If upstream migration is impeded, eel density could be higher below dams than in other areas and is expected to lead to decreases in growth (R. Verdon, pers. comm.). Richkus and Whalen (1999) considered that lack of information made it difficult to quantify both the overall direct and indirect impact of dams.

Resource managers have rated the effect of dams as one of the most significant potential causes of the yellow eel population decline along the Atlantic coast. Where recruitment has been impeded by an unmitigated barrier, a marked ecline in the upstream eel population has occurred. In the St. Lawrence River system, for example, the impedance of upstream passage and turbine mortality of downstream migrant spawners is a possible cause of the decline in recruitment and abundance of eels in that system. Population modelling has revealed that the effects of dams on recruitment along the Atlantic Coast was one of the three most influential factors. Oceanic effects on glass eel production and cumulative effects of fishing were the most important.

3.2.2 Habitat change

Barriers can create reservoirs that are potentially valuable eel habitat. However, habitat benefits only accrue when upstream passage is provided through installation of passage structures or partially mitigated by stocking.

Eel sex may be associated with habitat type (Section 3.3.1). Therefore, hydropower impoundments could contribute more females than the river in a natural state, and losses through turbine mortality may be compensated by production in the reservoir. An example of this has been documented in the upper reaches of the Ernes Lake system in north-west Ireland, which are characterised by low density of fast growing females (M. Matthews, pers. comm.).

3.2.3 Upstream fish passage facilities

The effect of barriers on upstream passage depends on how easily the structure can be ascended by the size of immigrating eel present (EPRI 1999). Successful upstream passage may be more likely close to estuaries than farther upstream, because small eels are better able to ascend over obstructions. Although eel are generally considered good climbers, vertical falls, even those a few centimetres high, and moderate to high velocity currents can prevent upstream migration of yellow eels (Porcher 1992).

The height of the structure, presence of seepage flow and surface roughness, often factors related to the age of the structure, can alter the scalability of a dam. This was clearly shown on the Richelieu River where the rebuilding of two old cribwork dams in the 1960’s led to a gradual decrease in silver eel landings since 1981. Monitoring of a pass on the upper dam suggested a blockage at the lower dam (R. Verdon, pers. comm.). Without knowledge of the type of structure present at each site, it is not possible to determine the total impact of installed dams.

Where barriers are impassable, ship locks can allow upstream passage. On the St. Lawrence River, the Beauharnois lock allows significant passage. A reduction in ship movement through the locks has been correlated with a decrease in upstream passage at the fish ladder at an upstream fish ladder (R. Verdon, pers. comm.).

Where upstream passage is a concern, cost effective facilities can be installed to mitigate the blockage. Traditional fishways designed for other species have not proven ineffective for eels (J. Therrien, pers. comm.), but cost-effective upstream passage facilities for upstream migrating eels have been installed in Europe and to some extent along the U.S. Atlantic Coast. A simple eel ladder with a 1.1m wide entrance on the Chambly Dam on the Richelieu River has allowed passage of 57% of the eels present at the dam face after one year of operation and close to 70% after two years (R. Verdon, pers. comm.).

3.2.4 Monitoring of juvenile recruitment

Counts of eels ascending the eel ladder on the Moses-Saunders Dam on the St. Lawrence River were carried out between 1974 and 1999. These data were used as a recruitment index of daily passage into Lake Ontario (Casselman et al. 1997). The peak recruitment occurred in 1982 (27.489 passed per day in the peak period) but was down to 27 eels in 1999, a drop of three orders of magnitude. This index is one of the primary lines of evidence for a recruitment decline at the extremity of the species range.

37 Barriers increase access to elvers for transfer and recruitment monitoring. The efficiency of the system installed and the monitoring methodology may change over time. Any changes to these aspects of operation must be carefully documented and the resultant bias must be quantified at the time of change.

3.2.5 Downstream passage

The passage of emigrating silver eels must be available so that reservoirs can contribute to spawning escapement. Where dams are associated with power generation, reported turbine induced mortalities have ranged from 6% to 37% (Richkus and Whalen 1999). Studies indicate that mortality increases with the size of eels and is highest for small turbines with high head (Monten 1985; Larinier and Dartiguelongue 1989).

Females silver eels are more vulnerable than the smaller males to turbine mortality during passage through power stations. Females are common inland and likely to encounter turbines upon spawning emigration. Hydro-power stations have the potential to have a greater impact on females, and effort should be concentrated on protecting these large eels.

Where multiple dams are present on a river system, turbine mortality is cumulative. A 26.5% mortality rate has been reported for the Moses-Saunders Dam on the outlet of Lake Ontario and another 18% on a second dam of the St. Lawrence. Together, these structures are responsible for three quarters of the total mortality (passage + harvest). Any modification that reduces turbine mortality (e.g., modification of turbines or a change in operating regime) would lead to an increase escapement for American eel from Lake Ontario.

The effectiveness of protective measures for downstream passage is poorly understood. Mitigation, such as spilling and shut down, is possible where runs of silver eel can be predicted and correlated, for example, with rainfall events or sudden drops in barometric pressure or temperature (Boubée et al. in press; A. Haro pers. comm.; C. Durif pers. comm.). Where downstream migration triggers are less well defined or unknown, other protective measures will be required. The effectiveness of behavioural measures such as sound or light barriers are unclear, and further research is recommended to determine the feasibility of these approaches.

Although ideally safe downstream passage of emigrant eel should be provided at large hydro dams, there is as yet no cost-effective means of protecting eels from entraining at large dams. In these cases compensation mechanisms such as protection of downstream habitat and transfer of juveniles to under-populated habitat with free downstream access could be considered.

Entrainment and turbine losses are not a problem at all dams. However, if free flow over the dam is not available during the downstream migration period, emigrants will be unable to progress downstream. In these situations, delay in downstream passage may lead to potential loss of migration synchronisation. Loss of condition as the result of the delay may compromise the ability of eels to reach spawning grounds and reproduce successfully.

3.2.6 Facilities and management action needed to ensure sustainable fisheries

Once a passage problem has been identified, a means of surmounting the barrier should be considered. Where the upstream habitat is suitable and downstream passage is not a major issue, an effective means of allowing the upstream passage for the size of eels that are blocked should be provided.

There is a need to document and describe cost-effective technology allowing upstream passage. European and New Zealand experience indicates that systems that include climbing media suitable for the range of eel sizes present can be effective for low head dams (less than 10 m). For higher dams, lift and trap and transport should be considered. Where multiple dams are present or where numbers of juvenile eels are simply not available (for example because of other effects downstream) mitigation in the form of trap and transfer may be considered. It is important to recognise however that the strong migration behaviour exhibited by juvenile eels may lead them to accumulate at the next upstream barrier.

3.3 Other Anthropogenic Effects

3.3.1 Physical habitat change

Human actions over the past 250 years have significantly reduced the amount of habitat available to eel. This loss of habitat would have significantly reduced the spawner biomass. Anthropogenic blockages (dams and weirs) have resulted in an overall 84% decrease in available habitat to American eel in the U.S. Atlantic states (Busch et al. 1998; Table 3.2.1.1). In the North Atlantic, the reduction in river length with free access has been reduced by 85%. For the Mid Atlantic area, the reduction is 88%, the South Atlantic 82%, and the Great lakes area 82%. The extent to which

38 these blockages are a barrier to upstream migrating eels is unknown. Lost production due to blockages cannot be estimated because eel density in the areas surveyed is unknown.

The alteration of riverine habitat for power production, or the creation of lakes for human recreation or the enhancement of other fish species, may change the sex ratio of emigrating spawners. There is evidence that particular sexes are associated with different types of habitat in some systems. In some Maine systems, freshwater stream habitat is associated with escapement of male silver eels, while lakes produce almost exclusively female silver eels. Females and males may have different habitat requirements for optimal biomass, and the reduction of available habitat for one sex could reduce the amount of spawner biomass produced and change the sex ratio of emigrating adults.

3.3.2 Chemical habitat change

3.3.2.1 Contaminants

Due to their benthic habit, long life, and high fat content eels are particularly likely to accumulate contaminants (Couillard et al. 1997). However, the impact of contaminants on the survival and reproductive success of this species is poorly understood (Couillard et al. 1997; ASMFC 2000). Eels in the St. Lawrence River showed pathological liver lesions associated with contamination levels of organochlorine compounds, but ovarian development was not affected (Couillard et al. 1997). Eels are known to persist in areas with high pollution levels (e.g., Hudson River, St. Lawrence River) and are a potentially valuable sentinel species for pollution monitoring. However, pollutants have been shown to affect the survival, growth and reproductive success of other fishes. In addition, Anguillid species may not be as able to adapt to pollution as other species of fish, due to their panmictic reproduction which does not allow for selection of locally-adoptive pollutant-resistant genes.

3.3.2.2 Productivity

Change in productivity is often associated with change in growth, generation time and fecundity in fishes. In the Hudson River, for example, eels of the same size tended to be older in fresh compared to brackish environments. Changes in productivity in the Great Lakes could affect the Lake Ontario-St. Lawrence River eel stock. Female fecundity should be maximised with maximum size and is compromised when early maturation at a small size occurs. Females appear to prefer estuaries, lakes and large rivers that are productive. At least in small high gradient streams of lower productivity, males predominate. This pattern suggests that changes in productivity could influence sex distribution and sex ratio.

3.3.3 Biological habitat change

3.3.3.1 Introductions

Game fish: Stocking of game fish into areas where eels occur is common in many areas. Management sometimes favours the proliferation of other species in habitat where eels occur through stocking of game fishes. If such stocking results in competitive interactions for food and habitat, it could negatively impact eel growth. This could have implications for the size at spawning and the number of years needed to produce a spawner. Alternatively, direct predation by the introduced species on eel, especially on the young of year and juvenile stages, would reduce survival and thus the number of spawners emigrating years later.

Stock enhancement with young eels: There is no significant use of glass eel and elver stocking into North American systems as a means of enhancing natural populations. The apparent declines in escapement and recruitment discussed elsewhere in this paper have only recently come to the attention of Canadian and the American fisheries managers, and stocking could be considered as a viable solution to stock decline in the future. Stocking should always be governed by the principles of adaptive management and accompanied by an evaluation program, so that effects can be measured and practices changed as necessary. In considering stocking as a management option, various issues must be considered as follows:

• Stock transfers can result in the spread of parasites, diseases and non-native organisms which could negatively impact eels or other species in the recipient water body. • Consideration should be given to the possibility that stocking might mask declines in “wild” recruitment, leading to potential complacency over the true status of a wild stocks. Regular assessment of mixed stocks and wild fisheries, in conjunction with documented stocking levels, might enable tracking of stocked cohorts within a system.

39 • The existing biodiversity of systems should not be compromised by inappropriate introductions of American eels. It should not be assumed that eels should be abundant in all waters, any more than any other species. There may be cases where other species take preference, or where eels would be an unwanted predator. • When population density is high, eels may tend to become male. High densities of juveniles have been associated with more males in European eel (Helfman et al. 1987), but the limited examination of this issue in Anguilla rostrata has not produced a firm prediction of sex determination with density (Helfman et al. 1987). Krueger and Oliviera (1999) reported evidence that high density favours production of males. However, there was an inverse relationship between male abundance and density across broad regions of the United States Atlantic coastal states (Kleckner and McCleave 1985). If there is an effect of density on sex determination, then stocking of young of year eels could impact the sex ratio and resultant spawner biomass. Increased intraspecific competition may be expected to reduce the growth of resident eels.

Swim bladder parasite: Anguillicola crassus is a nematode parasite of Anguillid eels. It originated in the Japanese eel (Anguilla japonica) and has been documented in the European eel (A. anguilla) and recently in American eel (A. rostrata)(Barse and Secor 1999). A. crassus has been implicated in acute mortality as well as internal injury and growth impairment. Part of its life cycle occurs in the eel swim bladder, and its departure through the swim bladder wall can cause injury and scarring. These effects on the swim bladder could impact a silver eel’s ability to travel to the Sargasso Sea spawning grounds and thus its reproductive success. The long-term impacts of A. crassus on Anguillid species are unknown.

Zebra mussel: The invasion of the Great Lakes region by zebra mussel has resulted in greatly increased water clarity. This transparency could impact the available habitat, eel behaviour and consequent growth.

3.3.3.2 Changes to spawning habitat

Sargassum is a large macroalgae which floats in large masses in the Sargasso Sea where eels spawn. It provides habitat to many fish, birds, and reptiles and is a major source of energy in this vast area. This plant is harvested for fertiliser and animal feed and its exploitation is unregulated. Harvest of Sargassum weed in the Sargasso sea has potential to degrade the amount and quality of spawning habitat and impact A. rostrata early life history (ASMFC 2000). The harvest of Sargassum in U.S. Economic Zone is to be phased out upon direction of the South Atlantic Marine Fisheries Council.

40 Table 3.1.1.1 Estimates of natural and fishing mortality in American eels. Mortalities giving rate per unit time are instantaneous. Mortalities given as percentages are finite (equivalent to exploitation rate). Site Stage Natural Fishing Natural Fishing mortality mortality mortality + mortality + emigration emigration rate rate East River, Chester, Nova Scotia Elver 0.0612-0.0675 day-1 30.8-32.1% Petite Trinité Yellow 0.09 - 0.73 yr-1 0 St. Lawrence estuary Silver 20-25% Prince Edward Island Yellow/silver 0.5 yr-1 0.26 yr-1 -1 -1 Chesapeake Bay, Maryland Yellow 0.25 yr 0.25 yr

Table 3.1.2.1 Comparison of instantaneous disappearance rates of eels in exploited and unexploited waters of Prince Edward Island, and calculation of fishing mortality. Exploited sites Unexploited sites Fishing Location Disappearance Location Disappearance mortality rate rate* Tidal waters 0.78 Freshwater ponds 0.25 0.54 Pinette estuary, 1973 0.73 Pinette estuary, 2000 0.28 0.45 Tidal waters 0.78 Pinette estuary, 2000 0.28 0.51 Mean 0.76 0.26 0.50 *Equivalent to natural mortality + emigration rate

Table 3.1.2.2. Assumptions of the life table model used to estimate effects of fishing mortality on egg production. Natural mortality and emigrationNatural mortality + emigration = 0.26, which is the mean disappearance rate calculated for rate unexploited PEI sites. It was assumed that emigration begins at age 8, and that natural mortality and emigration contribute equally to disappearance. Lorenzen's (1996) general mortality function was applied to weights-at-age derived from an age-weight von Bertalanffy curve for the Boughton River system. This gave mortalities of 0.52 to 0.43 for ages 6 to 16. The mortality from the Lorenzen function was multiplied by an adjustment factor, which was varied until the mean natural mortality for ages 8-16 was 0.5 x 0.26. The adjustment factor required to do this was 0.27.

Elver fishery None Directed fishery for silver eels None Mortality between emigration From Lorenzen's (1996) general weight function (unadjusted) and the spawning ground Fecundity Barbin and McCleave's (1997) weight-based fecundity formula (based on data collected in Maine). Yellow/silver eel fishing 0 in unexploited trial, 0.50 in exploited trial. mortality Recruitment to the fishery 50.8 cm, or age 6+

41 Table 3.1.4.1 Counts and estimated recruitment of juvenile eels to the Petite rivière de la Trinité. Date Total of nightly counts 1982 1983 1984 1985 1993 1994 1995 1996 1999 16 Jun 12 20 Jun 87 24 Jun 164 28 Jun 595 57 2 Jul 138 1,402 110 6 Jul 32 1,135 104 48 10 Jul 1,203 49 682 41 27 14 Jul 575 48 602 1,072 208 18 Jul 358 185 181 163 84 22 Jul 163 116 263 47 26 Jul 475 154 257 22 30 Jul 335 97 334 389 19 3 Aug 328 35 234 820 10 7 Aug 160 20 84 618 10 11 Aug 9 9 137 217 6 15 Aug 75 18 126 8 19 Aug 111 14

Seasonal total 4,576 4,389 1,046 1,117 3,681 763 4,047 5,547 933 Ratio count total:1999 count total 4.9 4.7 1.1 1.2

Estimated run 61,308 58,802 14,014 14,965

Mean of the best 10 counts 458 439 105 112 85 Ratio mean of the best 10:mean of the best 10 counts in 1999 5.4 5.1 1.2 1.3 Estimated run 66,979 64,242 15,310 16,350 Notes: From 1982 to 1996, three counts per night were made at 21:00h, 22:00h and 23:00h every 4 nights. Individual counts for 1982 to 1985 are not avalaible. In 1999, entries are the total of the catch + observed eels on the same rock at the same hours. The estimate of eels recruiting in 1999 (mostly <110 mm, 1+) was 12,500.

42 Table 3.2.1.1. Historic and current potential eel habitat in the Eastern United States (Busch et al. 1998).

Region Watershed Name Historical length Current length Lost length Number of (km) (km) (km) dams North Atlantic St. John River Basin 11335 148 11187 37 Penobscot River Basin 15245 207 15038 75 Maine to Kennebec River Basin 9186 208 8978 97 Connecticut Androscoggin River Basin 4467 195 4272 95 Maine Coastal - St. Croix 10884 5166 5718 98 Saco, ME, NH, MA 9414 1685 7729 212 Merrimack River Basin 11006 10 10996 533 Connecticut River Basin 20874 99 20775 941 MA - RI Coastal Area 7886 1589 6297 708 Connecticut Coastal 10335 1188 9147 713 St. Francois River Basin 850 1 849 13 Total 111482 10496 100986 3522 Mid Atlantic Richelieu Basin including 9126 1 9125 235 Lake Champlain drainage New York Upper Hudson 22389 1 22388 660 through Lower Hudson - Long Island 7781 1431 6350 519 Virginia Delaware Coastal Area 26934 5148 21786 1068 Susquehanna River Basin 52331 251 52080 684 Upper Chesapeake 14884 8862 6022 157 Potomac River Basin 28140 3281 24859 443 Lower Chesapeake 37727 5559 32168 884 Total 199312 24534 174778 4650 South Atlantic Chowan-Roanoke Coastal 36775 3632 33143 371 Neuse-Pamlico Coastal 23324 12452 10872 445 North Cape Fear Coastal 20471 5990 14481 626 Carolina Pee Dee Coastal 35880 6139 29741 1034 to Florida Edisto-Santee Coastal 41504 7003 34501 1942 Ogeechee-Savannah Coastal 34604 4508 30096 1028 Altamaha-St. Marys Coastal 37172 4673 32499 1353 St. Johns Coastal 82334 6582 75752 40 Southern Florida Coastal 8044 4893 3151 105 Total 320108 55872 264236 6944 Great Eastern Lake Erie 113 66 47 4 Lakes Southwestern 8076 1827 6249 67 Lake Ontario New York and Southeastern 16156 2877 13279 159 Ontario to Lake Ontario Québec Lake Ontario- 5740 622 5118 225 St. Lawrence Total 30085 5392 24693 455 Grand Total 660987 96294 564693 15571

43 400 800 700 Exploited tidal waters Pinette Estuary and unexploited Exploited 300 Exploited 600 freshwater ponds Unexploited 1973 500 Exploited: N = 2,315 N = 557 Unexploited: N = 3,652 200 400 300 Number of eels Number Number of eels Number 200 100 100 0 0 0 20 40 60 80 100 14.0 39.4 64.8 90.2 Length (cm) Length (cm)

10 Figure 3.1.2.1. Length frequencies of eels in exploited Middle and and unexploited waters of Prince Edward Island, Canada. 8 South Pinette Estuary 6 2000 Unexploited eels 4 N = 55 Number of of Number

2 800 Exploited tidal waters 0 Instantaneous 0 102030405060708090100 600 disappearance 0.784 Length (cm)

400 Actual 8 Simulated Middle & South Number of eels of Number 200 Pinette, unexploited 6 2000 0 Instantaneous 13579111315 disappearance Age 4 0.279 eels Number of 2

200 0 Pinette estuary, 1 3 5 7 9 11 13 15 150 exploited Age 1973 Instantaneous 500 100 disappearance Freshwater ponds, eels 0.726 400 unexploited Number of Number 50 Instantaneous 300 disappearance 0.246 eels 0 200 13579111315 Number of Age 100

0 13579111315 Age

Figure 3.1.2.2. Actual and simulated age frequencies of eels in exploited and unexploited waters of Prince Edward Island. Actual age frequencies are derived from length frequencies, using a von Bertalanffy age-length function.

44 4.0

Unadjusted 3.0 Lorenzen Adjusted Lorenzen 2.0

1.0 Natural mortality Natural

0.0 1 3 5 7 9 111315171921 Age

Figure 3.1.4.1 Instantaneous natural mortalities of eels on the Petite rivière de la Trinité, according to the unadjusted general fish equation of Lorenzen (1996), and the Lorenzen equation adjusted to make survival from age 1 to age 20 equal to 2%.

45 4 OPTIONS FOR DEVELOPING ESCAPEMENT TARGETS FOR SELECTED SYSTEMS

4.1 Evaluation of Human and Natural Impacts on American Eel Recruitment

In coastal fisheries uncertainty in the outcome of management is often the result of incomplete knowledge on life cycles and key demographic parameters. Resource management of American eel epitomises this dilemma: because spawning eels have not been sampled, there is no direct means to relate spawning stock to recruitment in a conventional manner (e.g., Beverton and Holt 1957), nor are there strong theoretical grounds to relate spawning stock to recruitment. Nevertheless, in the face of this uncertainty, the Precautionary Approach directs managers to take action to insure against future harm to the resource, making use of the best available understanding of life history attributes and vital rates.

In this section, the use of a Paulik diagram (Paulik 1973) serves to summarise key elements of the life cycle of eels and classify possible population-level consequences of natural and anthropogenic impacts. The Paulik diagram, originally developed for Pacific salmon, depicts a series of transition functions among critical life stages and provides a framework to evaluate population responses to single or multiple impacts. The Paulik diagram may be particularly useful for life cycles where key stages are thought to be regulated by density dependent processes (Nash et al. 1996).

The American eel life cycle can be separated into two broad phases: a) An oceanic phase in which silver eels migrate to the Sargasso Sea, spawn, and die; eggs and leptocephali drift; and glass eels disperse into near-shore habitats of North America. Very little is known about this phase, and what is known is largely outside of the influence of management. b) An estuarine/freshwater phase in which glass eels and elvers disperse throughout watersheds, and grow into a sedentary yellow eel stage. For this phase there is probably sufficient attainable knowledge to recommend action to maintain populations within limits imposed by recruitment and local productivity.

It is reasonable to suppose that the oceanic phase is driven principally by density independent controls whereas in the estuarine/freshwater phases, density-dependent processes can predominate, with production limited by availability of suitable habitat or food. These two phases are further subdivided into four key life cycle transitions:

1. silver eel production → egg production.

2. egg production → glass eel/elver (“pre-settlement”) production.

3. elver production → yellow eel production.

4. yellow eel production →silver eel production.

The first two transitions occur principally in the ocean, the latter two correspond to estuarine/freshwater phases. In a counter-clockwise rotation, four plots are combined leading from and returning to silver eel production (Figure 4.1.1, see legend for additional information on interpreting diagram). A summary of possible impacts for each transition is presented in Table 4.1.1.

4.1.1 The Paulik diagram

Silver eel → Egg transition: American eel life history differs substantially across latitude, habitat, and gender leading to widely differing maturation schedules. Following metamorphosis, all silver eels leave freshwater and estuarine habitats, initiating long-distance oceanic migrations (> 1000 km) to the Sargasso Sea where spawning occurs from February to April (Schmidt 1922; McCleave et al. 1987). Spawning has never been directly observed. American eels spawn small eggs (c. 1 mm diameter) and fecundity increases exponentially with length: 40 cm and 80 cm long silver eels yield 1 X 106 and 8 X106 eggs, respectively (Wenner and Musick 1974; Barbin and McCleave 1997).

Overall, egg production should increase as a linear function of silver eel production. Because the Sargasso Sea is vast, no regulating mechanism seems likely.

Depensatory effects of low silver eel production on egg production (shown by dotted line in Figure 4.1.1) can be envisioned for scenarios leading to poor mating success. If for instance, the sex ratio becomes skewed towards females

46 at low densities of estuarine/freshwater stages (Kreuger and Oliveira 1999; Oliveira 1999), then insufficient males may be present to ensure fertilisation. Alternatively, if spawning aggregations are disrupted at very low spawner densities, then the probability of mating and fertilisation in the Sargasso Sea, an enormous spawning habitat, could be impaired.

Egg glass eel/elver transition: Larvae (leptocephali) of the American eel are transported passively by oceanic currents, particularly the Gulf Stream, adjacent to the North American continental shelf. Larvae metamorphose into unpigmented juveniles (glass eels) and glass eels transit across the continental shelf. Dispersal from the Sargasso Sea to nearshore environments requires 6-16 months. Glass eel enter nearshore coastal habitats, estuaries, and usually migrate into non- tidal fluvial habitats. Dispersal into upper reaches of watersheds may continue after the glass eel has become a pigmented juvenile (elver). At some point however, the elver settles into the benthos and adopts a more sedentary life style. For convenience, the transition between egg and elver production only includes the pre-settlement dispersal period.

The transition between egg and elver stages is primarily influenced at both oceanic/decadal and continental shelf/seasonal scales that affect larval dispersal and juvenile ingress into estuaries. Effects are stochastic and because these operate in open systems, the conversion of eggs to elvers is modelled according to linear function with wide confidence intervals. As glass eels move into shallow coastal and estuarine habitats, it is possible that piscine and avian predation control regional elver abundance, perhaps in a density dependent manner. Still, abiotic controlling processes are expected to exert dominant control on this transition.

Depensatory effects of low egg production on elver production (shown by dotted line in Figure 4.1.1) might occur if successful dispersal of larvae is dependent upon their distribution or concentration. At low spawner densities, resulting larvae distributed in relatively few patches might not successively be entrained into oceanic/tidal currents that are likely to bring them to continental shelf/nearshore waters. If dispersal of larvae to continental shelf areas is concentration dependent (i.e., ideal free distribution as proposed by Smogor et al. 1995), ingress would be impaired.

Elver Yellow eel transition: The distinction between elvers and yellow eels is rather arbitrary, as there is no clear morphological or physiological difference between the stages. Here we designate elvers as pigmented juveniles <10 cm. Elvers and yellow eels are sedentary, showing strong affinity to local regions (100 – 10,000 m) in nearshore coastal environments, estuaries, and non-tidal fluvial and lacustrine habitats. Precise site affinity (“homing”) within these stages has been demonstrated through telemetry, tagging, and mechanistic studies. These stages show high rates of phenotypic plasticity in growth, sex differentiation, maturation size and age, diet, morphology, and habitat shifts. Many of these life history attributes are associated with habitats that are colonised by elvers/yellow eels. For yellow eels captured from major estuarine systems, Helfman et al. (1987) showed an inverse latitudinal trend in growth rate and a positive latitudinal trend in size and age at maturation. While Helfman’s (1987) proposed sex ratios also showed a latitudinal trend–increasing females with increasing latitude – more recent data (Krueger and Oliveira 1999) indicates that many males are produced at the northern part of the species range. Recently, Oliveira (1999) has provided additional evidence for Tesch’s original proposal (Tesch 1977) that sex ratio is most likely determined by density during the elver or early yellow eel stage. Recent analyses of otolith microchemistry indicate that yellow eels show different modes of habitat use – stage-long residency in freshwater, estuarine or coastal waters, and habitat shifts from freshwater to brackish water. These modes of habitat use have consequences to growth and maturation rates. Otolith microchemical analysis should allow habitat use during the yellow eel phase to be hind-cast from silver eels and thereby allow silver eel production to be related to broad categories of freshwater, estuarine, and coastal habitats (Secor and Rooker 2000).

Because site-affinity and density dependent processes are well documented for elver/yellow eel stages, compensatory functions were chosen to represent the transition between elvers and yellow eels. Territoriality, space and forage limitation, predation, and cannibalism are conceived as possible mechanisms leading to either a saturating or an over- compensating function. Either of these forms results in high degrees of regulation linked to habitat.

Yellow eel silver eel transition: Yellow eels undergo fundamental physiological and morphological changes during their metamorphosis to silver eels. Pigment and swim bladder morphology change, eye diameter increases, and the digestive system degenerates. Associated with these trends is recruitment of lipid and proteins to gonads. Male silver eel metamorphosis occurs at significantly smaller sizes (<40 cm) and younger ages than for females (>50 cm). Female silver eels are substantially larger and older in the northern part of the species range than in the southern part. During and after metamorphosis, silver eels initiate migrations out of watersheds and into coastal environments.

The transition to silver eels from yellow eels must be energetically costly. Additional energy reserves must be sequestered to undertake long distance ocean migrations and fully ripen eggs prior to spawning. Growth during the years leading up to metamorphosis must therefore control the transition. Because growth is likely regulated by density during the yellow eel stage, the functional transition is again modelled as either a saturating or an over-compensating function.

47 4.1.2 Evaluation of impacts

Functional relationships proposed for major life history transitions are speculative but permit likely consequences of single or multiple impacts to be tracked through the entire life cycle. Initially, it’s useful to consider an unimpacted population (Figure 4.1.2.1). A given level of silver egg production can be converted to egg production, elver production, etc. When the Beverton-Holt model of compensation is input during elver→silver eel transitions, then the deterministic functions alone would lead to a stable production of recruits. Alternatively, over-compensation transitions during the estuarine/freshwater phases should lead to an oscillating abundance level, symmetrical about some carrying capacity for these environments (Figure 4.1.2.1).

Next consider the loss of spawners to a low level, under a system of depensation at either the silver eel→egg transition or the egg glass eel/elver transition. Tracking the consequences of these functions through the life cycle shows a downward spiral that would lead to extinction (Figure 4.1.2.2). Impacts that could lead to reductions in silver eel production include exploitation and dam passage (see Section 3). Impacts that could reduce egg production include Anguillicola crassus parasitism, pollution, and altered sex ratios. The effects of reduced egg production can be exacerbated by oceanographic conditions unfavourable for larval survival and dispersal to continental shelf waters (Figure 4.1.2.2).

The effect of increased silver eel escapement on recruitment of elvers and yellow eels will be heavily conditioned by the expected compensatory functions. Thus proportional improvements of recruitment from increased spawner escapement should not be expected. Still, an effect of increased freshwater/estuarine habitat (increased passage, improved habitat, stocking into unoccupied habitats) can result in improvements to production of subsequent generations recruitment of elvers and yellow eels (Figure 4.1.2.3).

4.1.3 Spawning escapement and the Precautionary Approach – life history insights

There are important caveats to the presented Paulik framework related to complexity in age structure (overlapping generations), phenotypic plasticity, environmental sex determination, and high spatial heterogeneity in elver/yellow eel production. Given these limitations, the evaluation of impacts supports two findings:

1) The possibility of depensation at low spawning stock (Liermann and Hilborn 1997; Ludwig 1998) supports management for some minimum level of spawning stock escapement. While it seems unlikely that recent American eel spawning escapement has approached this level, within ICES there is concern that the more highly exploited European eel may be showing signs of a downward spiral in elver recruitment, perhaps due to the combined effects of low escapement and poor oceanographic conditions (ICES 1998). An analytical approach for determining escapement targets is presented in the next section. 2) Strong regulation during the estuarine/freshwater stages indicates that increases in spawning escapement may result in increased F1 elver recruitment, however this increase is unlikely to result in proportional changes in recruitment in subsequent generations. Habitat increase in estuaries and freshwater environments (e.g., increased passage) is viewed as a principal means to improve recruitment of elvers and yellow eels in current and future generations.

4.2 Biological Reference Points and the Precautionary Approach

ICES has recognised that a precautionary approach should be applied to fishery management and that reference points are key concepts in its implementation (ICES 2000). These reference points could be stated in terms of fishing mortality rates or biomass with the intention of ensuring that the stocks and their exploitation remain within safe biological limits. Implicit in the development of reference points is the assumption that there is a relationship between spawning stock and recruitment. The precautionary approach, dictates that unless it can be scientifically demonstrated otherwise, such a relationship between stock and recruitment should be assumed to exist (ICES 1997/Assess:7).

4.2.1 Reference points

The value of establishing reference points depends upon the consequences to the resource of variations in spawning stock abundance. There are two hypotheses to consider in deciding whether spawning stock reference points are appropriate for the American eel; recruitment dependent on spawning stock size versus recruitment dependent upon environmental conditions.

48 Consequences of managing for spawning stock size to future recruitment dependent upon the factor regulating recruitment Factor regulating recruitment Management approach Spawning stock Environment Ignore spawning stock size Risk of crashing the stock Variable and unknown rate of recruitment Manage for spawning stock size Reduced risk of crashing the Variable and unknown rate of stock recruitment The prudent action under the conflicting hypotheses is to minimise the risk of crashing the stock. This would be achieved by assuming dependence of recruitment on spawning stock size, consistent with a Precautionary Approach.

There are two general classes of reference points:

1) Limits: set boundaries which define safe biological levels. Limits are often referred to as thresholds and are intended to minimise the risk of the stock falling below a minimum size (Mace 1994; ICES 1997/Assess:7). 2) Targets: are reference levels to aim for and are intended to meet management objectives such as achieving yields close to the maximum attainable level (Mace 1994). Target reference points would be more conservative than limit points. Target mortality rates would be lower than the limit mortality rates whereas target spawner biomass reference points would be higher than limit spawner biomass levels (Figure 4.2.1.1). The management strategy would be designed to avoid exceeding the threshold and if the threshold is exceeded, then substantial reductions in mortality, including restrictions or prevention of the activity causing the mortality (for example fishing, turbine mortalities) would be considered (Rosenberg et al. 1994).

There exists clear guidelines for the establishment and application of reference points (ICES 1997/Assess:7, p. 4):

1) A reference point is an estimated value derived through an agreed scientific procedure. 2) Both limit reference points and target reference points should be used. 3) Management strategies shall ensure that the risk of exceeding limit reference points is very low. A large number of reference points (both mortality rates and biomass levels) and their associated data needs are summarised in ICES 1997/Assess:7 (Table 4.2.1.1). The majority of reference points require information on several population parameters including age structure, growth, natural mortality, spawning stock size and recruitment size. The interpretation of some reference points relative to a theoretical stock and recruitment relationship is shown in Figure 4.2.1.1.

The fishing mortality rate which generates maximum sustainable yield should be regarded as a minimum standard for limit reference points (ICES 1997/Assess:7). To be consistent with a precautionary approach, limits should be defined in terms of mortality rates and spawning biomass levels (ICES 1997/Assess:7).

There are advantages and disadvantages to the establishment and application of mortality rate limits and spawning biomass limits (Rosenberg et al. 1994).

49 Mortality limits Spawning biomass limits Advantages Relate directly to the activity that can Biomass is directly linked to be controlled recruitment Can be estimated from relatively Provide a guide for management of limited data and information on life stocks that are already depleted history characteristics Can prevent stock depletion due to the Provides a seed stock for eventual long-term activity recovery when adverse environmental conditions constrain abundance

Disadvantages Do not provide protection for stocks Difficult and extensive data to collect which are already at low level May require modification if Risk of mis-estimation when a limited environmental conditions and life range of stock conditions is available history characteristics change May be mis-interpreted as the point at which the resource will collapse

The management strategies could involve mortality rate limits and spawner biomass limits applied singly or in combination (Figure 4.2.1.2):

• fixed harvest rate policy: removals represent a constant proportion of the recruitment, • fixed escapement policy: all recruitment in excess of the spawning requirement is removed, • floor policy: a fixed proportion of the recruitment is removed when recruitment exceeds the minimum spawning requirement. The management strategies have consequences to yield and to spawning escapement. Use of fixed harvest rate policies has the least consequence on yield, i.e., fisheries are never closed but it has the most risk to the stock because the stock continues to be exploited even at low abundance. Continued harvesting at low abundance can have a destabilising effect which increases the risk of resource depletion (Beddington and May 1977). A fixed escapement policy has important consequences on yield because of fishery closures at low stock abundance but reduces the risk to the stock because at low abundance levels, spawners are protected. Floor policies have the greatest consequence to yield and the least risk to the stock. Lande et al. (1997) suggested that for a large class of population dynamics and a range of biological optimisation criteria (risk of resource collapse), the optimal strategy always involved a threshold population size above which all excess individuals are harvested. Thresholds are a necessary feature of harvesting strategies with the objective of minimising risks of resource depletion while optimising yields in variable environments (Lande et al. 1997).

4.2.2 Consequences of uncertainty

The greater the uncertainties, the greater the need to be precautionary (ICES 1997/Assess:7). Increased uncertainty renders optimal harvesting strategies more conservative and optimal threshold increases (Lande et al. 1997). Flim and Blim are reference points which should be avoided with high probability. There are uncertainties in the estimation of Flim and Blim as well as uncertainties in the assessments of the resource status relative to population abundance and exploitation. As a consequence of uncertainty, ICES (2000) defined precautionary reference points (Fpa and Bpa) to constrain exploitation to ensure a higher probability of not exceeding the limits. Fpa and Bpa are the main devices in the ICES framework for setting advice (ICES 2000).

4.2.3 Proposed limit reference points in data-poor conditions

The majority of reference points require information on several population parameters including age structure, growth, natural mortality, spawning stock size and recruitment size. The limited knowledge and particular population dynamics of American eel are a major obstacle to the derivation of reference points (Section 4.1). The wide distribution along the east coast of North America results in important differences in growth rates, age at maturity, and sex ratios. The mechanisms determining sex differentiation of animals are uncertain (growth rate, density, temperature, or a combination of factors). It is unclear how recruitment to freshwater occurs and whether there are regional stock and recruitment linkages. More importantly, there is little to no information on carrying capacity of habitat types for eels or what habitat variables determine carrying capacity. Natural mortality rates would vary with age and are likely high for the early life stages and decreasing with age and size.

50 Following the advice of ICES 1997/Assess:7, under data poor conditions, a mortality rate which provides 30% of the virgin (F=0) SPR is a reasonable first estimate of Flim until further information is gathered. Considering uncertainties, a preliminary estimate for Fpa could be 50% SPR.

Estimates of spawning stock and recruitment for the American eel are not available. In the absence of such data, ICES 1997/Assess:7 suggested that biomass index series such as CPUE series or survey-based measures could be used to establish relative Blim reference points. For example, the maximum survey index could be used as an indicator of virgin biomass and Blim would be some value of that maximum level, such as 20% of max. The estimate of Bpa could be set at a value higher than Blim, i.e. 50% of the maximum of the index series. At this time, no relevant indices of abundance of American eel exist on a continent-wide area. Consequently, no biomass reference points are proposed.

4.2.4 Preliminary reference points for American eel

Preliminary mortality rate reference points could be established across the entire species range. Any reference points established should consider the following:

• Given the difficulties in estimating and forecasting stock size, fishing mortalities should remain below M (natural mortality) (Walters and Maguire 1996),

• Uncertainty in estimated population size increased Blim (Lande et al. 1997), • Ability to monitor compliance.

Mortality rate reference points across the species distribution would ensure levels of escapement proportional to abundance in each stock area. Overall loss to spawning stock depends upon the number of years eels are vulnerable to the fishery. Reference exploitation rates would vary with region – higher in the south than in the north.

Preliminary values of Flim were derived from a spawner to recruit analysis and provisionally determining Flim at the F which generated 30% SPR. Fpa was estimated from the 50% SPR profile. These mortality reference points are estimated for eels aged one year and older. They could be calculated to include the elver stage but consideration for density- dependent regulation at the elver to yellow and silver eel stage would have to be considered (see Section 4.1).

The maturation schedule of eels is not well known, but female silver eels in northern areas are on average older than those in the south. The same is true for male eels. Some representative maturation schedules were examined to see the effect of these on the SPR solutions (Figure 4.2.4.1). Simple partial recruitment vectors considered in the example calculations assumed full and constant recruitment at a given age (age 3 years or age 12 years in the examples). This is a likely situation in some of the fyke net fisheries in tidal waters in eastern Canada which operate throughout the ice-free season, but would not be the case in other areas (Section 6.1).

An example calculation estimating Flim and Fpa for eel with variable maturation schedules probably typical of northern area and southern area stocks is shown in Figure 4.2.4.2. The %SPR function is relatively insensitive to the natural mortality assumption (as seen by the width of the crescent profile) for the northern area assumptions but was more important for the southern area. Flim to Fpa range was narrow (between F = 0.06 and 0.12) for the northern area stock and wider (between F = 0.11 and 0.32) for the southern area stock (Figure 4.2.4.2). The maturation schedule is particularly important in the estimation of Flim and Fpa because this determines the number of years the animal is exposed to the fishery.

The reference points are also sensitive to the partial recruitment vector assumption. The partial recruitment profile would respond to management actions such as size limits on retained eels, mesh size limits, area restrictions, and seasons. In the second example, the effect of different partial recruitment vectors (fully recruited at age 3 years versus fully recruited at age 12 years) (Figure 4.2.4.1) but for a fixed maturation schedule (northern profile) is described. The Flim and Fpa points increase as the age of full recruitment to the fishery increases. Flim increased from 0.09-0.12 at fully recruited age 3 to 0.14-0.18 at fully recruited age 12 (Figure 4.2.4.3). Fpa increased from 0.06-0.07 at fully recruited age 3 to 0.08-0.118 at fully recruited age 12 (Figure 4.2.4.3). The higher reference point levels correspond to the reduced availability of animals to the fishery over their residence time in continental waters.

4.2.5 Limitations of Flim method

The estimates described above are based on equilibirum conditions, i.e. no change in characteristics with abundance. Adding stock and recruitment to the model has an effect on yield calculations, i.e. yield declines with increasing spawning stock size (Hilborn and Walters 1992). Defining only Flim reference levels can be dangerous because an F-

51 based definition appropriate over a middle range of biomass levels may not be appropriate at the extremes of biomass. Also, the definitions set to prevent long-term decline of the stock do not increase the protection to the resource when it is in poor condition (Rosenberg et al. 1994).

4.2.6 Application of Flim and Fpa

The estimation of Flim and Fpa levels applicable to a stock are dependent upon information on the age and size composition, maturation schedule, and characteristics of the fishery itself including size selectivity and availability of life stages to the gear.

When the mortality factors on the stock are managed such that F equal to or less than Fpa, there should be a low probability that the realised mortality is not sustainable (ICES 1997/Assess:7). In the absence of Blim and Bpa reference points, other measures of stock status would be used to assess compliance with the limit and PA points. These indicators would include size composition of the catch relative to unexploited areas, relative abundance of yellow and silver eels (when these are available for capture), condition factors, etc.

52 Table 4.1.1. General Life Cycle of American eel Anguilla rostrata with possible control and regulatory mechanisms.

Transition Function Control s (impacts) Regulation silver eel → egg Fecundity- Pollution Mating system (sex ratio) deterministic Anguillicola crassus egg→ Oceanographic- Climate-ocean currents Unlikely except in near shore glass eel/elver stochastic Tides, gyres environments (pre-settlement) Predation Foraging success Sargassum harvest Oceanographic- As above Minimum larval production needed for depensatory dispersal to North American watersheds elver → yellow Beverton-Holt Predation Predation eel saturating Passage Dispersal (post-settlement) Habitat fragmentation/loss Foraging success Pollution Territoriality Exploitation Cannibalism Climate/Ecosystem effects Ricker- As above Increased importance of overcompensatory 1.Interference competition (space, food) 2. Cannibalism yellow eel Beverton-Holt Passage Dispersal →silver eel saturating Climate/Ecosystem effects Foraging success Pollution Anguilllicola crassus Ricker- As above Interference competition overcompensatory

53 Table 4.2.1.1. Commonly used reference points and their associated data requirements (source: ICES 1997/Assess:7).

Reference Point Definition Data requirements Fishing rates F0.1 Fishing mortality rate (F) at Weight at age, natural mortality, which slope of the yield per exploitation pattern (F, partial recruit (Y/R) curve is 10% of its recruitment vector) value near the origin Fmax F giving the maximum yield on a Weight at age, natural mortality, Y/R curve exploitation pattern (F, partial recruitment vector) FMSY F corresponding to Maximum Weight at age, natural mortality, Sustainable Yield from a exploitation pattern (F, partial production model or from an age- recruitment vector), stock based analysis using a stock recruitment relationship or recruitment model general production models F30%SPR F corresponding to a Spawning Weight and maturity at age, Stock Biomass per Recruit natural mortality, exploitation (SSB/R) which is 30% of the pattern (F, partial recruitment SSB/R obtained when F = 0 vector) F >= M Empirical (for top predators) M and sustainable F’s for similar resources Fcrash F represented by the tangent Weight at age, natural mortality, through the origin of a stock exploitation pattern (F, partial recruitment relationship recruitment vector), stock recruitment relationship Fpa F precautionary approach, used to Same as data required for other F constrain mortality to ensure high reference point calculations probability of exceeding Flim (mortality limit reference point) Biomass / Spawning Stock Levels BMSY Biomass corresponding to Weight at age, natural mortality, Maximum Sustainable Yield exploitation pattern (F, partial from a production model or from recruitment vector), stock an age-based analysis using a recruitment relationship or stock recruitment model general production models MBAL A value of SSB below which the Data series of spawning stock probability of reduced size and recruitment recruitment increases B50%R The level of spawning stock at Stock recruitment relationship which average recruitment is one half of the maximum of the underlying stock recruitment relationship B20%B-virg Level of spawning stock Weight at age, natural mortality, corresponding to a fraction (for exploitation pattern (F, partial ex. 20%) of the unexploited recruitment vector), stock biomass recruitment relationship Bpa B precautionary approach, used Same as data required for other B as constraint on mortality to reference point calculations ensure a high probability of exceeding Blim (biomass limit reference point)

54 Eggs

0 Glass eel/Elver Silver Eel

Yellow Eel

Figure 4.1.1. Paulik diagram of life cycle of American eel Anguilla rostrata. Critical life history transitions are represented for production rates of four stages: silver eel, egg, glass eel/elver (presettlement juvenile stages), and yellow eel. Null production for each stage occurs where axes intersect. Production of stages proceeds in a counter-clockwise manner from silver eels →eggs→elvers → yellow eels → silver eels. For the silver eel-egg transition, a dotted line presents scenario of depensation. Dotted lines also present an alternative scenario of density dependent regulation during elver-yellow eel and yellow eel-silver eel transitions. Dashed lines for the egg-glass eel/elver transition indicate the possible range and stochasticity of oceanographic effects on elver recruitment.

55 Egg

0 Glass Silver

Yellow

Egg

0 Glass Silver

Yellow

Figure 4.1.2.1. Production trajectory among life stages of American eel Anguilla rostrata. Production trajectory of stages proceeds in a counter-clockwise manner from silver eels →eggs→elvers → yellow eels → silver eels. Top panel and bottom panel show trajectories according to Beverton-Holt and Ricker transition functions in estuarine/freshwater habitats.

56 Eggs

0 Glass eel/Elver Silver Eel

Eggs Yellow Eel

0 Glass eel/Elver Silver Eel

Yellow Eel

Eggs

0 Glass eel/Elver Silver Eel

Yellow Eel

Figure 4.1.2.2. Production trajectory among life stages of American eel Anguilla rostrata. Production trajectory of stages proceeds in a counter-clockwise manner from silver eels →eggs→elvers → yellow eels → silver eels. Oval highlights portion of Paulik diagram corresponding to low silver eel production. Top breakout panel shows scenario of low spawning escapement, followed by a depensatory response. Bottom breakout panel shows combined scenario of early depensation and poor oceanographic conditions (shown by bold dashed line).

57 Eggs

0 Elver Silver Eel

Yellow Eel

Figure 4.1.2.3. Production trajectory among life stages of American eel Anguilla rostrata. Production trajectory of stages proceeds in a counter-clockwise manner from silver eels →eggs→elvers → yellow eels → silver eels. Scenario is shown for increased estuarine/freshwater habitat (bold dotted line in lower left hand quadrant).

58 FTarg FMSY Fcrash

MSY

B MSY BTarg Relative Recruitment Abundance Relative Recruitment 0 0 Relative Spawning Stock Abundance

Figure 4.2.1.1. Position of some mortality rate and spawning biomass reference points relative to a theoretical stock recruitment relationship. The reference points are described in Table 4.2.1.1.

59 Fixed FMSY Fixed BMSY Floor(BMSY&Ftarg) Relative Yield

BMSY 0 0 Relative Spawning Stock Abundance

BMSY

Fixed FMSY Fixed BMSY Floor(BMSY&Ftarg) BMSY Relative Spawning Escapement Relative Spawning

0 0 Relative Spawning Stock Abundance

Figure 4.2.1.2. Effect on yield (upper panel) and on spawning escapement (lower panel) based on management policies using reference points. Fixed FMSY refers to managing on a fixed fishing mortality rate set at the fishing rate which generates MSY. Fixed BMSY refers to managing on a fixed spawning escapement (harvesting all recruitment in excess of BMSY) set at the spawning escapement which generates MSY. Floor refers to managing using BMSY and Ftarg when recruitment exceeds BMSY. The yield and spawning escapement plots were calculated using the theoretical stock and recruitment relationship in Figure 4.2.1.1.

60 1.0

0.8 M min M max Maturity schedule-A Maturity schedule-B 0.6 Relative fecundity PR Vector (age = 3) Rates PR Vector (age = 12) 0.4

0.2

0.0 0 5 10 15 20 25 30 Age (years)

Figure 4.2.4.1. Input assumptions to the spawner to recruit modelling to estimate Flim and Fpa. Maturity schedule refers to the proportion of the potential female yellow eels destined to metamorphose to silver eels. PR vector refers to the partial recruitment vector to the fishing gear. Maturation schedule A refers to a northern area stock and schedule B would be representative of a southern area stock.

61 100%

90% M = 0.05 to 0.40 80% PR: fully recruited at age = 3 Maturation profile: northern 70%

60%

50%

40%

30%

% Spawning Per Recruit (SPR) Recruit Per % Spawning 20% Fpa F 10% lim

0% 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Rate of Fishing Mortality (F)

100%

90% M = 0.05 to 0.40 PR: fully recruited at age = 3 80% Maturation profile: southern 70%

60%

50%

40%

30%

% Spawner Per Recruit (SPR) Recruit Per % Spawner 20%

10% Fpa Flim

0% 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 Rate of Fishing Mortality (F)

Figure 4.2.4.2. Estimated %SPR relative to F for the eel stock of the northern area (upper panel) and a southern area (lower panel) for varying assumptions of M. The estimated %SPR is eggs per R (adjusted for fecundity at length).

62 100%

90% M = 0.05 to 0.40 80% PR: fully recruited at age = 3 Maturation profile: northern 70%

60%

50%

40%

30%

% Spawning Per Recruit (SPR) Recruit Per % Spawning 20% Fpa F 10% lim

0% 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Rate of Fishing Mortality (F)

100%

90% M = 0.05 to 0.40 80% PR: fully recruited at age = 12 Maturation profile: northern 70%

60%

50%

40%

30%

% Spawning Per Recruit (SPR) Recruit Per % Spawning 20%

Fpa 10% Flim

0% 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Rate of Fishing (F)

Figure 4.2.4.3. Estimated %SPR relative to F for varying assumptions of natural mortality (M) for the northern stock area with fixed maturity profile but different partial recruitment vectors.

63 5 MANAGEMENT UNITS FOR ESCAPEMENT TARGETS

5.1 Application of the Precautionary Approach

Genetic evidence suggests that the American eel is panmictic (Williams et al. 1973; Keohn and Williams 1978; Williams and Keohn 1984; Avise et al. 1986; 1990). The passive drift of eel leptocephali also suggests a panmictic population. Consequently, the number of recruits entering a river in a given year should not be related to the number of spawning eel that left the river in previous years, but should be related to the total number of eel from all geographic areas that spawned successfully. And to oceanographic conditions that affect their geographic distribution.

The most conservative application of the Precautionary Approach (see Section 4.2) to the development of management units for American Eel has several aspects. The Precautionary Approach would suggest that if recruitment declines were detected anywhere in the range, the most appropriate management would be to increase escapement of eel throughout the range. The Precautionary Approach would further suggest that, in case there are distinct geographic populations of American eel populations which have not been detected, escapement of eel should be increased more dramatically in the area where recruitment declines are detected.

5.2 Management Units

Substantial differences in life history parameters (growth, maturity schedule, sex ratio, etc.) exist within the eel’s range e.g., Helfman et al. (1987), Facey and Van Den Avyle (1987). However, substantial differences in these parameters also exist within a single region and in some cases within a single river, resulting in no clear biological basis for identifying management units.

Geographic units are needed to provide a workable basis for management actions (e.g., increased escapement). The Working Group felt that management of eel would ideally occur on a primary watershed basis. This watershed unit would include all fisheries and other anthropogenic impacts that occur on an eel stock and would also assist in the maintenance of genetically distinct populations in the event that the species was found not to be panmictic.

The watershed approach poses two difficulties for implementation: 1) many small watersheds exist for which no information on eel is available to fisheries managers, and 2) in large watersheds (e.g., the St. Lawrence River) several fisheries management jurisdictions may be involved in the management of one eel stock.

The Working Group resolved that management units should be defined by the existing jurisdictions for eel management with the caveat described in Section 5.3. These jurisdictions include the Atlantic Coastal States in the United States of America, the Department of Fisheries and Oceans Canada, the Province of Québec, and the Province of Ontario.

5.3 Multi-Jurisdictional Eel Stocks

In areas where eels migrate through several jurisdictions during their life cycle the jurisdictions involved should cooperate to meet the management objectives for this eel stock. For example, in the St. Lawrence River, the Provinces of Québec and Ontario and the State of New York should cooperate on the management activities pertaining to the eels from the upper St. Lawrence River and Lake Ontario stock. Other areas where cooperation on eel management will be required include the Chesapeake Bay, the Hudson River, and the Delaware River.

5.4 Outside of Canadian/US Jurisdiction

American eel utilize a large geographic range including the east coast of North America as far north as Labrador. In addition, eel have been documented on the east coast of Central America, the islands of Caribbean Sea, and north- eastern South America, but little is known about their fisheries. Additional information should be collected on these fisheries if they exist and the relevant fisheries management agencies encouraged to cooperate with actions to increase eel escapement (Section 7).

64 6 MANAGEMENT ACTIONS FOR REQUIRED ESCAPEMENTS

With reliable data on recruitment, catches, effort, escapement, and the status of stocks it would be possible to consider long-term management, define reference points for fishing mortality and spawning stock biomass and coordinate management efforts across the geographic range of the American eel. However, the current data-starved position requires a pragmatic approach before such facts and figures are available.

In relation to fishery controls, there are various management options potentially available to fishery managers in Canada and the United States to achieve the reference points recommended in Section 4. Management jurisdictions within these two countries already have, to various degrees, management plans for the species in place as outlined below in Section 6.1. Section 6.2 outlines some of the more commonly used management options that may be implemented or strengthened to achieve the management goals.

6.1 Current Management Plans

Canada: Management plans for the American eel in Canada are implemented by the respective provincial governments in Ontario and Québec and by the federal government in the Maritime Provinces and Newfoundland (Table 6.1.1). The plans vary according to province and within New Brunswick and Nova Scotia, by the major watersheds, i.e., the Gulf of St. Lawrence (designated Gulf) and the Atlantic coast of Nova Scotia plus the Bay of Fundy (Scotia-Fundy). Only the provinces of Ontario and Québec exploit a common geographical segment of the eel population through their respective fisheries in Lake Ontario and in the St. Lawrence watershed.

Most of the catch (probably greater than 95 %) of American eel in Canada is taken by commercial fisheries, with the remainder harvested by recreational and aboriginal fisheries. Catch statistics for these latter two sectors are sparse.

United States: Prior to the development of the Atlantic States Marine Fisheries Commission (ASMFC) Fisheries Management Plan (the Plan), commercial eel fisheries were managed by the States according to regulations provided in Table 6.1.1. At its November 1999 annual meeting, ASMFC voted to approve the first Interstate Fishery Management Plan for American Eel (Anguilla rostrata) to be effective January 1, 2001. The Plan is a working document that describes the goals and objectives for the species, its current status, recent and historical trends, the ecological challenges affecting the species, management options and actions needed to reach and maintain the goals, and issues that need additional research support. In general, an ASMFC species management plan provides for the regulation of human activities that impact a species so that the population remains sustainable and viable, while also supporting the natural diversity of the ecological system(s) it inhabits.

The specific management unit for the Plan is defined as that portion of the American eel population occurring in the territorial seas and inland waters along the USA Atlantic coast, including all States and jurisdictions from Maine to Florida. The Plan also recognises that significant numbers of eel inhabit areas outside of ASMFC jurisdictional boundaries. These include watersheds in the Canadian Atlantic Provinces, upstream freshwater reaches that are managed by inland fish and wildlife agencies of ASMFC member States, and regional institutions such as the Gulf States Marine Fisheries Commission, as well as those waters within Native American Reservations where Tribal Governments have jurisdiction. Effective management of American eel in USA waters needs to recognise the interests of applicable Governments/agencies outside its jurisdiction, including nations both north and south of the U.S. to implement holistic methods in the protection and enhancement of this species.

Implementation of the ASMFC management program for American eel will be based on scientific advice provided by scientific staff of the state and federal agencies, as well as input from public hearings and an ASMFC Citizen’s Advisory Panel. In general, management will strive for a long-term sustainable population, with a surplus to support recreational, subsistence, and commercial fisheries. Each State is responsible for implementing management measures and protecting American eel habitat within its jurisdiction to ensure the viability of the population segment residing within its boundaries.

Recreational Fisheries Management Measures: Currently there is an observed, but undocumented recreational fishery for American eel within the Atlantic coast States. The harvest rate is unknown, as is the discard mortality rate of American eel from recreational fisheries for other species. In order to minimise the chance of excessive recreational harvest, as well as circumvention of commercial eel regulations, states are required to establish a framework management program for recreational fishers, consisting of a 6-inch uniform size limit, possession limit not to exceed 50 eels, with no sale permitted by recreational fishers.

65 Commercial Fisheries Management Measures: For commercial fisheries, States shall institute licensing and reporting mechanisms to ensure that annual effort (including total units of gear deployed) and landings information by life stage (glass eel/elver, yellow eel, and silver eel) are provided by harvesters and dealers. Further, States are required to maintain existing or more conservative regulations, including gear specifications contained in Table 6.1.1, for all life stages. States with minimum size limits shall retain those size limits, unless otherwise approved by the ASMFC American Eel Management Board.

Management Measures in Federal Waters: The ASMFC American eel FMP further recommends that the U.S. Secretary of Commerce address and initiate controls over harvest and use of American eel in federal waters (3-200 miles offshore of the U.S. coast) that are not landed in states. Specifically, the Plan recommends that the Secretary of Commerce ban harvests of American eel at any life stage in the U.S. EEZ, but permit possession of up to fifty (50) eel per person for use as bait.

International Trade: In addition to existing channels for documenting exports, it is also recommended that the Secretary of the Interior proceed with listing American eel glass eels and elvers under Appendix III of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). An Appendix III listing in no manner prohibits the harvest of American eel at any life stage. The Appendix III listing will improve the regulation, protection, and control of glass eel in the lucrative, but largely undocumented international trade. The listing provides for monitoring and inspection at the port of departure and also at the port of arrival at the importing country through the use of a permit system. A CITES Appendix III export permit indicates that a legal harvest has taken place in accordance with the permit issuing authority.

6.2 Management Options

Possible management options which are discussed in more detail below include measures to limit exploitation by fisheries, protect and improve the productive capability of eel habitat, and enhance production through expansion of accessible habitat and the stocking of under-seeded or inaccessible habitat.

6.2.1 Measures to limit exploitation by fisheries

Measures to limit exploitation by fisheries will most often be site/area and circumstance specific; they will generally function by regulating the length of time that an individual eel will be possibly affected by the fisheries.

Prohibition of fishing: Prohibition of fishing can be life stage specific or area specific; for example, some areas (e.g., Pennsylvania) prohibit all commercial fishing, while others prohibit elver fishing or fishing in rivers. Many watersheds where eel production exists do not currently have fisheries (e.g., parts of Newfoundland and north shore of Québec) and, based on the conclusions of Section 2 on trends in recruitment and fisheries, it would be prudent to prohibit the introduction of new fisheries.

Total allowable catches/quotas: Ideally, application of total allowable catch/ quota restrictions requires knowledge of abundance and identification of escapement targets. Quotas put an upper limit on the total catch; however, with the diverse nature of the eel fisheries, it is difficult to envision how an individual quota on a panmictic stock would be shared and subsequently managed/enforced. Required data are not currently available for different life-stages of eel and therefore TAC approaches are probably not workable.

Gear controls: Controls on, for example, number, size, mesh-size, usage, and location of gear are already enforced in most American eel fisheries to control fishery mortality (Table 6.1.1). Where they do not exist, they should be introduced and in other areas strengthened.

Landing size limits: Minimum size restrictions could help to reduce excessive exploitation of yellow and pre-spawner eel. For example, minimum mesh size limits could be imposed for fyke nets and eel pots. Limits on maximum size would promote escapement of larger (that is: female) pre-spawners, but could also trigger increases in fishing effort aimed at smaller sizes.

Closed seasons: Closed seasons are currently used in some areas. They are commonly based on traditional or practicable fishing season or are primarily related to requirements to allow unhindered migration of other species such as salmonids. The effectiveness of fishing time controls is affected by temporal variations in eel activity and migrations, often as a result of changing environmental parameters. The timing of closed seasons must be related to local characteristics of eel and fisheries and has to primarily consider closure during periods of vulnerability. Some jurisdictions have implemented closed seasons for eel fishing unrelated to their vulnerability.

66 Closed areas: These could be locally effective, e.g., in preventing extension of fisheries (particularly for glass eel/elvers) into new areas or for protection of vulnerable glass eel/elver or silver runs. Alternatively, closed areas could be used to designate ‘reserves’ or ‘refuges’ where no exploitation would be permitted; such an approach is currently used in the management of eel stocks in New Zealand and could be applied to many watersheds of Canada and the United States where unfished eel populations are known to exist.

Licensing for fishermen and dealers: Licensing specific to eel for all fishermen and fishing gear and dealer licensing could help provide, via catch returns and market statistics, improved information for monitoring catches and compliance with targets. The quality of such information is currently often poor, but licensing of fishermen and gear, in conjunction with adequate enforcement of regulations, offers opportunities for controlling and monitoring fishing effort and, ultimately, fishing mortality.

6.2.2 Upstream and downstream passage

The effects of dams is the most significant factor affecting the upstream and downstream passage of eels where the structures are impassable or are shown to cause a significant delay in migration. There has been a long history of the successful use of upstream passage facilities as a cost effective means of mitigating the effects of hydro dams in Europe, New Zealand, and elsewhere. There are few sites along the Atlantic coast of North America where effective upstream eel passage facilities have been constructed. There appears to be considerable variability in the effectiveness of facilities installed for the upstream passage of eels. An effective management action would be to provide standards for the construction of fish passage facilities to maximise efficiency. In particular there is a need to specify a standard substrate that allows the passage of a variety of sizes. It is recommended that fisheries management authorities develop a manual for the construction of fish passage ways for eels, and to identify characteristics of sites where these structures should be installed.

Trap and transport methods also provide for an effective means of mitigating the effects of dams, particularly where consecutive dams are in close proximity. It is recommended that this method of mitigating the effects of dams be considered when evaluating the needs for eel passage at specific locations.

The downstream passage of migrating silver eels past dams remains problematical. The documented mortality rates caused by turbines at hydro dams is between 18% and 26.5%, although a higher mortality may occur with small turbines or where the head is high. These mortality rates need to be included with fishing mortality rates to evaluate the overall impact on eel stocks.

There are several management options available to mitigate the impacts of dams on downstream passage. These include the implementation of water management practices that might allow the spilling/overflow of water during the migration period for silver eels. Where water intakes are present, the installation of screens to prevent entrainment should be installed, in conjunction with effective bypass systems. It is recognised that some of these mitigation methods will be impractical at some sites. It is recommended that fisheries managers consider mitigation methods for eel passage on a case-by-case basis. Evaluation should take into account that if there is increased production above dams provided by upstream passage facilities or stocking, then mortality rates caused by turbines may be offset by the increased production upstream.

6.2.3 Transfers and re-stocking

Under the correct circumstances, transfers and stocking of juvenile eels to growing areas can be useful in increasing freshwater production and, by inference spawner escapement. Re-stocking may hence be considered as an alternative or additional practice to reductions in fisheries where increased spawner escapement is a requirement. Transfer of growing stock may also be useful in restoring locally depleted stocks, for instance upstream of dams or other barriers. Short range stocking programs (i.e., within watersheds or very local regions) are considered to be very much less risky than those involving transfers between watersheds or over greater ranges, which may contribute less well to spawner escapement and carry a higher risk of disease transfer.

The following are some key concerns relating to stocking:

• There should only be a presumption in favour of stocking where there is likely to be a positive, or at least no negative influence on overall escapement, taking into account all human-induced mortality. For example, stock should not be transferred into waters where there are likely to be impediments to emigration, for example upstream of hydroelectric turbines where mortality is known to be high, or above dams impassable to downstream migrants.

67 • Due consideration should be given to biodiversity issues–it should not be assumed that eels should be abundant in all waters, any more than any other species. There may be cases where other species take preference, or where eels would be an unwanted predator. • The spread of parasites, diseases and non-native organisms that may be associated with eels should always be avoided. • Consideration should be given to the possibility that stocking might mask declines in “wild” recruitment, leading to potential complacency over the true status of a wild stock. One possible means of avoiding this would be regular and structured surveys of mixed stocked and wild fisheries, which, in conjunction with documented stocking levels, might enable tracking of stocked cohorts within the system. • Stocking should always be governed by the principles of adaptive management and accompanied by an evaluation program, so that effects can be measured and practices changed as necessary.

6.3 Conclusions/ Recommendations

Consideration must always be given to the fact that anthropogenic impacts on the American eel occur almost solely on prespawning stages of this animal and that those that survive all fisheries and downstream passage are destined to only spawn once.

Under a Precautionary Approach, considering the signals indicated in Section 2, it would be prudent to not allow any increased exploitation and to prohibit the development of any new fisheries.

It is recommended that fisheries management authorities provide a manual for the construction of fish passage ways for eels, and to identify structures where these characteristics of sites should be installed.

It is recommended that fisheries managers consider mitigation methods for eel passage on a case-by-case basis. Evaluation should take into account that if there is increased production above dams provided by upstream passage facilities or stocking, then mortality rates caused by turbines may be offset by the increased production upstream.

68 Table 6.1.1. Regulations governing eel fisheries in Canada and the U.S. Atlantic coast States in 2000.

Province/ Sector Waters Gear Open Licence Minimum Comments State season re- size (cm) quired that may be retained

Ontario commercial inland trap nets, year round for yes no size limit. Harvest restricted by limiting new licenses, individual quotas and hoop nets, hooklines, others closed seasons, and buy-out have occurred. Harvest has remained hooklines & seasonal well below quota. electro- fishing recreational all waters angling year round yes 20 No directed fishery.

Québec commercial St. Law- longlines year round yes 20 rence, (38), eel fresh- traps (150) water commercial pound net 1April - 30 Nov yes 20 (28), hoop net (~4000) commercial pound net year round yes 20 (24) commercial marine & pound net 1 Sept - 30 Nov yes 20 tidal (139) commercial inland all types closed yes 20 lakes & rivers Native food all waters all types year round no 20 No known fishery.

recreational all waters angling year round yes 20 No directed fishery.

Newfoundland commercial tidal & pots 1 June/1July - 31 yes 20 Re-issuance (transfer) of eel licenses is not permitted other than inland Oct through a transfer of complete enterprise. commercial fyke nets mid-Aug - 31 yes 20 Oct experimental head of nets 20 June - 31 Aug yes <10 Elver fisheries, nine licenses with fixed quotas. tide recreational tidal pots 1 June/1July - 31 yes 20 Oct recreational tidal spears yes 20

69 Table 6.1.1. (cont'd). Regulations governing eel fisheries in New Brunswick, Prince Edward Island, Nova Scotia and the U.S. Atlantic Provincet St t i 2000Secto Water Gear Open Licenc Minimu Comment /State seaso re- size quire ()that d be ti d New commerci tidal trap year yes 38.1 In 1996 the minimum size in all Gulf NB eel GulfBikl t d increasedfi h i from 20 to 38.1 commerci tidal pots year yes 38.1 This fishery is legally open, but is recreationl tidal spears 16 d Nov-31 no 38.1 iti Nlativ tidal trap yearA yes 38.1 t d New commerci tidal & pots, eel year yes 30 Commercial and recreational licenses frozen at Scotia-Bikl inland nets, set d Commercialtl l licenses transferable, recreational Fd lines & transferable.li Annual re-newal only to those longline plirevious d iyear. th Transfer of an existing commercial lic pt erson is only permitted to a person registered and whodi participated in a commercial fi h

recreation tidal pots & year yes 30 l longline d recreation inland &angling year no 30 l tidal, spearin& d except spearing only in tidal commerci head of traps, pots1Mar -31 yes < 10 Elver fisheries, two commercial licenses with fixed qu l tide dip& A quotas(li and individual river quotas). Elver fisheries t on riversitt d not having other eel Nativ tidal & all year yes Nfiative h i fisheries are licensed, either as food, social inand d or il as il Prince commerci tidal trap 16 Aug - 15 yes 50.8 A 46 cm minimum size in effect on all PEI eel fisher IslanEd d l t Ot 1970s,ith was raised to 50.8 cm 1998. In 1999, the o d the trap-netf fishery was changed from 16 Aug - 31 O 1516 A commerci tidal spears 17 May - 30 yes 50.8 SpearingOt takes place at night with the aid of generato l J lights (flambeauing).d Prior to 1993, the season was Aug.A15 Prior to 1996, spearing was also permitted in N theI1999 season was changed from 1 Apr - 30 Jun to 17 M 30 J

70 Table 6.1.1. (cont'd). Regulations governing eel fisheries in New Brunswick, Prince Edward Island, Nova Scotia and the U.S. Atlantic coast States in Province/ Sector Wat ers Gear Open Licence Minimum Comments State season re- size (cm) quired that may be retained

Prince Edward recreational tidal spears 1 Jan - 31 Mar no 50.8 Regulations specify a closing date of 30 Jun, but the season is Island closed on 31 Mar by administrative policy. Native tidal trap nets 16 Aug - 15 Oct yes 50.8 Native tidal spears 17 May - 30 Jun yes 50.8

Nova Scotia Gulf commercial tidal trap nets 1 Sep - 31 Oct yes 50 In 1997 the minimum size in all Gulf Nova Scotia eel fisheries was increased from 20 to 50 cm. commercial tidal pots 1 Sep - 31 Oct yes 50 recreational tidal spears 1 Sep - 31 Oct no 50

Nova Scotia commercial tidal pots, eel year round yes 20 Commercial and recreational licenses frozen at current level. Scotia-Fundy traps & Commercial licenses transferable, recreational licenses non- weirs transferable. Annual re-newal only to those licensed in the previous year. Transfer of an existing commercial licence to a new person is only permitted to a person registered in proceeding year and who participated in a commercial fishery.

commercial inland pots, eel 15 Aug - 31 Oct yes 20 traps & weirs recreational inland & angling & year round no 20 tidal, spearing except spearing only in tidal recreational tidal pots year round yes 20 experimental & head of dip nets, eel 1Mar -31 Aug yes < 10 Elver fisheries, three commercial licenses and five experimental commercial tide traps & pots licenses with fixed quotas (license quotas and individual river quotas). Elver fisheries only permitted on rivers not having other eel fisheries. Native tidal & all types year round yes Native fisheries are licensed, either as food, social and ceremonial inland or as commercial.

71 Table 6.1.1. (cont'd). Regulations governing eel fisheries in New Brunswick, Prince Edward Island, Nova Scotia and the U.S. Atlantic coast States in 2000. Province/ Sector Waters Gear Open Licence Minimum Comments State season re- size (cm) quired that may be retained Maine commercial tidal & fyke nets & 15 Mar - 6 June yes none Glass eel/elver fishery in effect, variable license fee by gear and by inland weirs combination of gears. Pot mesh size 1.27cm x 1.27cm. 6 June - 3 Mar 15.24cm

New commercial tidal pots & nets yes 15.24 cm Coastal netting license required for nets and pots. Hampshire

Massachusetts commercial tidal yes 10.16 cm Multiple license requirements.

Vermont commercial inland yes none $500 electrofishing + $25 per crew member; Lake Champlain

Rhode Island commercial tidal yes 15.24 cm No fishing within 1/2 mile of fish ladder outlet.

Connecticut commercial tidal yes 15.24 cm Glass eel/elver fishery in effect. Dip net glass eel 3/1-5/31 with weekly closed periods.

New York commercial tidal pots & nets yes 15.24 cm 2.54cm x 1.27cm pot mesh or escape panel. commercial inland weirs, pots yes none Pot entrance not > 5.08cm diameter. & set lines

Pennsylvania inland 15.24 cm Commercial fishng for eel prohibited.

New Jersey commercial tidal pots & fyke yes 15.24 cm Glass eel/elver fishery closed. Multiple license options/fees for nets yellow eel fishery.

Delaware commercial tidal yes 15.24 cm

Maryland commercial tidal pots yes 15.24 cm 1.27cm x 1.27cm pot mesh or escape panel, limited entry.

Potomic River commercial tidal pots yes 15.24 cm $75US per boat. Fisheries Com.

District of commercial tidal 15.24 cm Columbia

Virginia commercial tidal pots yes 15.24 cm Glass eel/elver fishery in effect, limited to two special permits. 1.27cm x 1.27cm pot mesh or escape panel, license plus gear fees, 2 year wait.

West Virginia commercial inland yes none Except 5/15-6/30, jigging,snagging, snaring are prohibited.

72 Table 6.1.1. (cont'd). Regulations governing eel fisheries in New Brunswick, Prince Edward Island, Nova Scotia and the U.S. Atlantic coast States in Province/ Sector Waters Gear Open Licence Minimum Comments State season re- size (cm) quired that may be retained

North Carolina commercial tidal pots yes 15.24 cm 2.54cm x 1.27cm pot mesh, 20 eel limit per person per day.

South Carolina commercial tidal pots & fyke yes none Glass eel/elver fishery in effect, dip nets licensed. Gear permit nets required in addition to State license, 1.27cm x 1.27cm pot mesh.

Georgia commercial tidal pots yes 15.24 cm 3.81cm x 1.27cm pot mesh.

Florida commercial tidal pots yes none Glass eel/elver fishery in effect, limited to 3 special permits. 2.54cm x 1.27cm pot mesh.

Note: U.S. Atlantic coast States regulations subject to change. Escape panels of varying sizes by State are required.

73 7 COORDINATION OF MONITORING AND RESEARCH

In sharp contrast to assessment information collected for species populations with a relatively narrow range, the widespread spatial distribution of American eel is such that truly representative monitoring may not be achieved given current practices. Indications are that many life history characteristics vary throughout the range of American eel. The scale of impacts to eel life history varies widely from localized to oceanic levels. A coordinated, international effort to collect relevant data would allow better management advice to be given than is currently possible.

Therefore, it is strongly recommended that an international commission be formed to organise monitoring and research. The commission would serve as a clearing house for regular exchange of information regarding landings and resource status, and it would provide insight on research needs. The commission would conduct periodic workshops to facilitate international exchange of information and to suggest modifications to research and monitoring programs as required, and provide advice on management.

7.1 Monitoring

Agencies involved in monitoring should meet to discuss standardisation of methods for monitoring elver recruitment and silver eel escapement. At the very minimum, current monitoring of all life stages should continue as currently designed, and any additional emphasis should be placed on quantitative, rather than qualitative results. Consideration should be given to a watershed approach in the design of glass eel/elver recruitment surveys. Monitoring of silver eel escapement should reflect management strategies designed to assess compliance with reference point calculations. Sufficient numbers of silver eel should be sampled to determine ambient sex ratios at each monitoring site. Monitoring of yellow eel abundance and size/age frequency is recommended to describe the nature of relationships between elver and later life stage abundance, and to monitor sources of mortality.

7.1.1 Current monitoring: recruiting life stages

Canada: Two monitoring surveys presently exist for glass eels/elvers. One survey has occur ed since 1996 in the East River, Chester, Nova Scotia the other began in Prince Edward Island in 2000. The first of these occur in the Scotia- Fundy region of New Brunswick and in Nova Scotia (East River). Additional elver/juvenile sampling has recently been initiated (1996) on the north shores of the St. Lawrence River, Québec.

USA: The Atlantic States Marine Fisheries Commission has instituted a glass eel monitoring program. Beginning in 2001, every participating Atlantic coast state is required to conduct an annual survey of glass eel.

7.1.2 Current monitoring: yellow eel

Canada: Yellow eel monitoring exists in the upper St. Lawrence/ Lake Ontario system at Moses-Saunders Hydroelectric Dam and through long-term commercial fishing operations and trawl sampling. Monitoring at the Moses- Saunders site is considered to be of paramount importance to the understanding of the American eel resource regionally and range wide.

USA: Limited monitoring of yellow eel abundance exists in various State waters, however the ASMFC Plan (ASMFC, 2000) recommends that a comprehensive monitoring plan be established to address yellow eel.

7.1.3 Current monitoring: silver eel

Canada: Commercial catch and effort monitoring have been conducted in the St. Lawrence River since 1979. Silver eel monitoring programs have been initiated in 1999 in the Petite Rivière de la Trinité and Sud Ouest River (Section 2).

USA: No coordinated programs to assess escapement of silver eel are currently being conducted. However, the ASMFC management plan (ASMFC 2000) recommends the formulation of a coast-wide sampling program for American eel using standardised and statistically robust methods.

74 7.2 Research

7.2.1 International coordination of research

As recommended earlier, an international commission should be formed to guide monitoring and research. Canada and the United States are obvious participants, however inclusion of representatives from the Gulf States region of the USA, as well as representation from Caribbean and Central and South American interests need inclusion in order to encompass the full geographic range as well as relevant issues regarding management of the species.

A task of the proposed commission would be to identify and tabulate a list of current and recommended research activities. It is suggested that a survey of current research is conducted, and a database is maintained by the proposed international eel commission, to track active research initiatives. The international commission should produce periodic reports of active research as recommended at the commission’s annual workshops.

7.2.2 Research priorities

7.2.2.1 Baseline landings and survey

Baseline landings and survey information is essential to proper management. Currently, total landings are not known with precision, nor can landings be differentiated by life stage, area (inland and coast), fishery type, or method in many places. Such information is essential to determine trends in recruitment, fisheries, stock abundance, and yield. Surveys of elver-yellow eel-silver eel abundance are critical to understanding the nature of the relationship between life stage dynamics.

In similar fashion, many basic life history parameters are unknown by life stage or location. Studies of basic eel biology including sex ratio, male and female growth rates, habitat preferences, predator/prey relationships; behaviour and movement during freshwater residency; oceanic behaviour, movement, location and spawning of mature eels; and examination of basic leptocephalus biology including nutrition, metamorphosis triggering mechanism, and mortality rate would be valuable in deciding proper management actions for the species.

Since eel inhabit freshwater and near-shore habitat for a large portion of their life span, research on the effects of pollution and accumulation or biological concentration of various xenobiotic contaminants on growth, maturity, and reproductive success would help provide insight into the management of the species.

Stock assessment and determination of current and sustainable fishing mortality rates and acceptable biomass levels are desirable aspects of any fisheries management plan, however such activities require well justified input parameters in order for resultant management advice to reflect biological reality and to resonate among the management community. In order for this concept to be realised for American eel, baseline data collections need to be improved throughout the range of the species.

7.2.2.2 Ecological evaluation of biological processes

Ecological evaluation of biological processes helps provide understanding from the empirical results of baseline studies. Such evaluations provide predictive value and insight towards population level response to management action. Latent population biology issues exist where determination of the existence of sub-populations or verification of panmixia requires further study. Continuation of studies on genetic differences between American and European eel should include evaluations of migratory routes and guidance mechanisms for silver eel in the ocean, factors affecting larval and juvenile survival, mechanisms of exit from the Sargasso Sea, transport across the continental shelf, and impacts on elver recruitment and distribution due to climate change and oceanic circulation.

Other necessary ecological information includes interactions of age, growth and natural mortality, particularly as confounded by sex, habitat, and possibly latitude. Primary needs involve age-at-entry of American eel glass eel into estuaries and fresh waters, mechanisms of sex determination, as well as evaluation of fecundity/length, fecundity/weight relationships from throughout the geographic range. Also, there is a need to determine natural mortality at specific life stages and within sizes of the yellow eel stage over the geographic range of the species. Determination of rates of maturation by cohort, and provision of a schedule of maturity is needed, as would be the development of an age/recruitment/migration chronology, using archived eel otoliths to determine cyclic patterns of eel abundance important to the understanding of oceanic influence and population trends.

75 Other ecological processes in need of evaluation include prey/predatory aspects of eel life history and their relative importance in the ecological energy flow of the systems they inhabit.

7.2.2.3 Research into anthropogenic impacts

Research into anthropogenic impacts on eel survival require evaluation. First, fishing losses, specifically those associated with fisheries for juvenile eel for use as bait, as well as impacts of elver fishing on abundance, distribution, and subsequent life history of later life stages within a watershed, need clarification. Second, as losses of American eel due to sources other than fishing are of concern, a database registry of barriers to passage, if available, would be helpful in determining impacts to ascending elvers due to habitat loss, or the degree of impingement/entrainment mortality for out-migrating silver eel. Continued research should 1) evaluate the impacts of upstream and downstream barriers to assess areas of extirpation and historical distribution, 2) model effects of hydroelectric facilities on eel population structure to provide a framework from which to judge the benefits of bypass improvement and mitigation options supporting key stock elements, and 3) evaluate downstream migration behaviour of eels through hydroelectric facilities as compared to migratory behaviour in un-impacted systems.

7.2.2.4 Research of basic methodological processes

Research of basic methodological processes as well as the exchange of ideas and experiences helps to improve results in baseline research. As such, the objective evaluation and validation of various age determination techniques, by an independent study group, should evaluate the degree of precision and accuracy associated with each method. Also, future research should strive to investigate, develop, and improve technologies for American eel passage upstream and downstream.

7.3 International Coordination of Management

Given that the suggestion has been to formulate an international commission to coordinate monitoring and research, logic suggests that this same group may be in a position to provide advice on international management. A management board of the proposed international commission could provide advice to those relevant nations on what, if any, agreements or alliances should be formed to manage shared resources. Three such modes of agreements become immediately obvious. The first being an agreement in North America dealing with American eel as a single species, with the relevant nations involved. The second being an agreement for multiple species which could involve several shared species in North America such as American eel, striped bass, American shad, and Atlantic sturgeon. The third mode of agreement could occur through either modification of the North American Salmon Conservation Organisation (NASCO) responsibilities to include European and American eel in addition to Atlantic salmon. Cooperation on eel management could be undertaken within NASCO, and separate North American and Northeast Atlantic commissions and NASCO’s relationship with ICES could be used to obtain regular scientific advice on Atlantic salmon and the two eel species.

Regular meeting of the management board would report on management strategies and proposed changes to management. Proposed changes would be derived through advice from technical committees, stock assessment committees, as well as advisory committees consisting of objective reviewers from various stakeholder groups including industry and academia.

76 8 CONCLUSIONS AND RECOMMENDATIONS

8.1 Conclusions

A review of available data confirmed either declining or neutral abundances of American eel in Canada/ USA and large declines in both recruitment of young eels to, and large fecund female silver eel escapement from the St. Lawrence River system which is situated near the northern end of the species range. Only one index demonstrated an increase in abundance. As well, there has been a major loss of habitat across the range and the possibility exists that oceanographic conditions may be affecting recruitment to the continent.

Eels are exploited at all life stages but the impact on overall silver eel escapement is unknown. Some areas in Canada/ USA remain unexploited but their contribution to escapement is unknown. Given that a panmictic hypothesis for the species has not been disproved, recruitment for northern areas is assumed to be dependent on total spawning stock size. Northern areas may also be more sensitive to larval dispersal mechanisms associated with oceanographic conditions.

Thus the Working Group concluded that reductions in habitat, declining or neutral trends in abundance, severe decline in abundance in northern areas, continuous exploitation and unknown oceanographic effects support the adoption of the Precautionary Approach in management.

8.2 Recommendations

Evidence from some areas of the American eel’s range suggests that reductions in human-induced mortality (which includes both fisheries and hydro dam turbine mortalities) of yellow and silver eels may be required. However, the data to develop escapement biomass limits are not currently available but need to be developed and implemented as soon as possible. Consistent with the Precautionary Approach, the Working Group recommends that there be no increased exploitation in areas where exploitation occurs, no development of fisheries in areas where there is currently no exploitation, and efforts be made to reduce human-induced mortality wherever possible.

Also, effort should be made to improve data gathering (particularly catch information in Canada and USA as well as in the Caribbean and Central and South America), monitoring compliance with existing regulations and to implement appropriate regulatory measures where none exist.

Attention should also be given to resolving fish passage problems at obstructions that are limiting access to production areas, and contributing to mortality of migrating silver eels. There is a need to document and describe cost-effective technology allowing upstream and downstream passage and to develop manuals for locating and constructing fish passage at obstructions. Consideration should also be given to the trapping and upstream transfers of eels and minimally, the transfer of juveniles to under-populated habitat with free downstream access.

The Working Group resolved that management units should be defined by the existing jurisdictions for eel management, and in areas where eels migrate through several jurisdictions during their life cycle, that those jurisdictions should cooperate in meeting the management objectives for the stock. In addition to the collection of fisheries and catch information from the Caribbean and Central and South America, the relevant management agencies should also be encouraged to adopt a Precautionary Approach to the management of the American eel.

In recognition that the American eel is panmictic, future management will require international coordination. Fishery scientists require an international forum in which to annually exchange information, guide monitoring and research, facilitate workshops to e.g., evaluate and validate various age determination techniques, and to develop management advice (see Section 7.). Better information is required on catches, exploitation, life history parameters and demography for determination of current and sustainable exploitation rates and acceptable escapement biomass limits. As well, there is an important need to develop estimates of carrying capacity of eel habitat throughout the species range and to establish long term monitoring programs of recruitment and spawner output.

77 APPENDIX 1

Working Documents Submitted to the Working Group on North American Eels, 2000:

Doc No. 1 Caron, F. Overview of eel situation and recent research in Québec. Is the American eel declining everywhere in the St. Lawrence?

Doc No. 2 Caron, F. G. Verreault and E. Rochard. Estimation of the numbers of American eels (Anguilla rostrata) leaving the St. Lawrence watershed and their exploitation rate.

Doc No. 3 Verreault, G., P. Pettigrew, R. Tardif, G. Pouliot and J.-F Gaudreault. The exploitation of the migrating silver American eel in the St. Lawrence River estuary, Québec, Canada.

Doc No. 4 Verreault, G., and P. Dumont. Escapement of American eel from the Upper St. Lawrence River and Lake Ontario.

Doc No. 5 Robitaille, J.A., P. Bérubé, S. Tremblay, and G. Verreault. Eel fishing in the St. Lawrence River/Great Lakes system during the 20th century: signs of overfishing.

Doc No. 6 Jessop, B.M. Catches of American eel, by State, for the Atlantic and Gulf coasts of the United States.

Doc No. 7 Jessop, B.M. The status of American eel in the Scotia-Fundy area of the Maritime Provinces as indicated by catch and licence statistics.

Doc No. 8 Jessop, B.M., and C. J. Harvie. An update and reanalysis of the EPRI CPUE data.

Doc No. 9 Jessop, B.M., C.G. Hannah, J.W. Loder and S.P. Oakey. An exploratory model-based study of elvers drift across the Scotian Shelf and Gulf of Maine.

Doc No. 10 Harvie, C.J. Watershed areas and stream lengths in Nova Scotia, Canada.

Doc No. 11 Beak International Inc. The decline of American eel (Anguilla rostrata) in the Lake Ontario/ St. Lawrence River ecosystem: a modeling approach to the identification of data gaps and research priorities.

Doc No. 12 Weeder, J.A., and J.H. Uphoff Jr. Effect of changes in growth and eel pot mesh size on American eel (Anguilla rostrata) yield per recruit in upper Chesapeake Bay.

Doc No. 13 Oliveira, K., J.D. McCleave, and G.S. Wippelhauser. Regional variation and the effect of habitat on sex distribution of American eels, Anguilla rostrata.

Doc No. 14 Cairns, D.K., C.D. MacPherson, and M.C. Lister. A preliminary life table for the American eel in the southern Gulf of St. Lawrence, Canada.

Doc No. 15 Cairns, D.K., and M.C. Lister. Landings, population densities, and population trends of American eels in the Maritime Provinces of Canada.

Doc No. 16 Cassleman, J.M. and L.A. Marcogliese. Synchronous changes in abundance of the upper St. Lawrence River-Lake Ontario stock of the American eel (Anguilla rostrata): evidence for general species decline.

Doc No. 17 ASMFC. Atlantic States Marine Fisheries Commission American eel young-of-the-year recruitment survey.

Doc No. 18 Rosell, R. and M. Allen. Understanding the dynamics of the Lough Neagh eel fishery.

Doc No. 19 Rosell, R. and T. Potter. A precautionary management framework for the European eel (Anguilla anguilla).

78 Doc No. 20 Secor, D., W. Morrison, J. Steinbacher, and J. Baker. Stock dynamics of American eels in freshwater and brackish habitats of the Hudson River, New York.

Doc No. 21 Dekker, W. Insufficient spawning stock is a sufficient explanation of the recruitment decline in the European eel.

Doc No. 22 Knights, B. Executive summary - eel stocks & management in England & Wales.

Doc No. 23 Jessop, B.M. Annual and seasonal variability in the size and biological characteristics of the runs of American eel elvers to two Nova Scotia rivers.

Doc No. 24 Chaput, G. Biological reference points for American eel.

Doc No. 25 McCleave, J. Decadal-scale trends in proxies for atmospheric and oceanic circulation in the North Atlantic Ocean.

79 APPENDIX 2

References Cited

Anthony, V. 1982. The calculation of F0.1: a plea for standardization. Northwest Atlantic Fisheries Organization, Serial Document SCR 82/VI/64, Halifax, Canada.

Atlantic States Marine Fisheries Commission (ASMFC). 2000. Interstate Fishery Management Plan for American Eel. Fishery Management Report No. 36. Washington, D.C. 79p.

Avise, J.C. G.S. Helfman, N.C. Saunders, and L.S. Hales. 1986. Mitochondrial DNA differentiation in North Atlantic eels: population genetics consequences of an unusual life history pattern. Proc. Natl. Acad. Sci. USA 83:4350- 4354.

Avise, J.C. W.S. Nelson, J. Arnold, R.K. Koehn, G.C. Williams, and V. Thorsteinsson. 1990. The evolutionary genetic status of Icelandic eels. Evolution 44:1254-1262.

Barbin, G.P., and J.D. McCleave. 1997. Fecundity of the American eel Anguilla rostrata at 45o N in Maine, USA. J. Fish Biol. 5:840-847.

Barse A.M., and D.H. Secor 1999, An exotic nematode parasite of the American eel. Non-native species management. Fisheries 24(2):6-10.

Beddington, J.R., and R.M. May. 1977. Harvesting populations in a randomly fluctuating environment. Science 197:463-465.

Beverton, R.J.H. and S.J. Holt. 1957. On the dynamics of exploited fish populations. Fisheries Investigation Series 2, Vol. 19. U.K. Ministry of Agriculture and Fisheries, London.

Boubée, J.A.I., C.P. Mitchell, B.L. Chisnall, and E.J. Bownman. (In press.) Factors regulating the downstream migration of mature eels (Anguilla spp.) at Aniwhenua Dam, Bay of Plenty, New Zealand. New Zealand J. Mar. Freshw. Res.

Busch, W.D.N., S.J. Lary, C.M. Castilione, and R.P. McDonald. 1998. Distribution and availability of Atlantic coast freshwater habitats for American eel (Anguilla rostrata). Administrative Report #98-2. USFWS, Amherst NY 28p.

Casselman, J.M., L.A. Marcogliese, and P.V. Hodson. 1997. Recruitment index for the upper H. Saunders hydroelectric generating station at Cornwall, Ontario, 1974-1995. pp 161-169 In Peterson, R.H. (Ed.). The American eel in eastern Canada: stock status and management strategies. Proceedings of eel workshop, January 13-14, 1997, Québec City, QC. Can. Tech. Rep. Fish. Aquat. Sci. No. 2196: v + 174p.

Castonguay M., P.V. Hodson, C.M. Couillard, M.J. Eckersley, J-D Dutil, and G. Verreault. 1994. Why is recruitment of the American eel, Anguilla rostrata, declining in the St. Lawrence River and Gulf? Can. J. Fish. Aquat. Sci. 51:479-488.

Couillard, C.M., P.V. Hodson, and M. Castonguay. 1997. Correlations between pathological changes and chemical contamination in American eels, Anguilla rostrata, from the St. Lawrence River. Can. J. Fish. Aquat. Sci. 54:1916-1927.

Dutil, J.-D., M. Michaud, and A. Giroux. 1989. Seasonal and diel patterns of stream invasion by American eels (Anguilla rostrata) in the northern Gulf of St. Lawrence. Can. J. Zool. 67:182-188.

EPRI. 1999. American eel (Anguilla rostrata) scoping study: A literature and data review of life history, stock status, population dynamics, and hydroelectric impacts. Electrical Power Research Institute Report TR-111873.

Facey, D.E., and M.J. Van Den Avyle. 1987 Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (North Atlantic) - American eel. US Fish Wildl. Serv. Biological Report 82(11.74) U.S. Army Corps of Engineers, TR EL-82-4. 37 p.

Fahay, M.P. 1978. Biological and fisheries data on American eel, Anguilla rostrata (LeSueur). U.S. Nat. Oceanic Atmosph. Admin. Northeast Fish. Center Sandy Hook Lab., Tech Series Rep. No. 17.

80 Fernandez, J. and J. Vasquez 1978. Las pesquerias de angulas en la Provincia de holguin [American eel (Anguilla rostrata , Le Sueur) fisheries in Holguin Province, Cuba]. Rev. Cub. Invest. Pesq. 3(2): 48-61.

Greeley, J.R. 1932. Fishes of the area. A biological survey of the Oswegatchie and Black River Systems. New York Conservation Department, Biol. Surv. 6:54-92.

Helfman, G.S., D.E. Facey, L.S. Hales Jr., and E.L. Bozeman, Jr. 1987. Reproductive ecology of the American eel. Am. Fish. Soc. Symp. 1:42-56.

Hilborn, R. and C.J. Walters. 1992. Quantitative Fisheries stock assessment: choice, dynamics and uncertainty. Chapman and Hall. 570p.

ICES 1997/Assess:7. Report of the Study Group on the precautionary approach to fisheries management. ICES 1997/Assess:7.

ICES 1998. European Eel. Extract of report of the Advisory Committee on Fishery Management No. 11. ICES, Copenhagen, Denmark.

ICES 2000. ICES cooperative research report No. 236, Report of the ICES Advisory Committee on Fisheries Management, 1999, Part 1.

Jessop, B.M. 1987. Migration eels in Nova Scotia. Trans Amer. Fish. Soc. 116:161-17

Jessop, B.M. 1997. An overview of European and American eel stocks, fisheries, and management issues. Pp. 6-20. In Peterson, R. H. (Ed.). The American eel in eastern Canada: stock status and management strategies. Proceedings of eel workshop, January 13-14, 1997, Québec City, QC. Can. Tech. Rep. Fish. Aquat. Sci. 2196: v + 174p.

Jessop, B.M. In press. Size and exploitation rate by dipnet fishery of the run of Ameriacanel (Anguilla rostrata) elvers in the East River, Nova Scotia. Dana 00:000.

Jessop, B.M. 2000. Estimates of population size and instream mortality rate of American eel elvers in a Nova Scotia river. Trans. Amer. Fish. Soc. 129:514-526.

Kelly, K.A. 1991. The meandering Gulf Stream as seen by the Geosat altimeter: surface transport, position, and velocity variance from 73° to 46° W. J. Geophys. Res. 96:12699-12709.

Kelly, K.A., and D.R. Watts. 1994. Monitoring of Gulf Stream transport by radar altimetry and inverted echo sounders. J. Phys. Oceanogr. 24:1080-1084.

Keohn R.K., and G.C. Williams. 1978. Genetic differentiation without isolation in the Amercian eel, Anguilla rostrata. Evolution 27:192-204

Kleckner, R.C., and J.D. McCleave. 1985. Spatial and temporal distribution of American eel larvae in relation to North Atlantic ocean current systems. Dana 4:67-92.

Knights, B. 1996. Risk assessment and management of contamination of eels (Anguilla spp.) by persistent xenobiotic organochlorine compounds. Chemistry and Ecology [Chem. Ecol.] 13(3):171-212.

Krueger, W., and K. Oliviera. 1999. Evidence for environmental sex determination in the American eel (Anguilla rostrata). Environ. Biol. Fish. 55:381-389.

Lande, R., B.-E. Sæther, and S. Engen. 1997. Threshold harvesting for sustainability of fluctuating resources. Ecology 78:1341-1350.

Larinier, M., and J. Dartiguelongue. 1989. La circulation des poissons migrateurs: le transit à travers les turbines des installations hydroélectriques. Bull. Français Pêche Piscicult. 312-313:1-90.

Liermann, M., and R. Hilborn. 1997. Depensation in fish stocks: a heirarchic Bayesian meta-analysis. Can. J. Fish. Aquat. Sci. 54:1976-1984.

81 Lorenzen, K. 1996. The relation between body weight and natural mortality in juvenile and adult fish: a comparison of natural ecosystems and aquaculture. J. Fish Biol. 49:627-647.

Ludwig, D. 1998. Management of stocks that may collapse. Oikos 83:397-402.

Mace, P.M. 1994. Relationships between common biological reference points used as thresholds and targets of fisheries management strategies. Can. J. Fish. Aquat. Sci. 51:110-122.

McCleave, J.D., R.C. Kleckner and M. Castonguay. 1987. Reproductive sympatry of American and European eels and implications for migration and taxonomy. Amer. Fish. Soc. Symp. 1:286-297.

Monten, E. 1985. Fish and turbines: fish injuries during passage through power station turbines. Vattenfall, Stockholm, 111p.

Nash, R.D.M, R. Berghan, A.D. Rijnsdorp, and H.W. Van der Veer. 1996. Exploration of the population dynamics of Irish Sea plaice, Pleuronectes platessa L. through the use of Paulik diagrams. J. Sea Res. 40(1-2):1-18.

Oliveira, K. 1999. Life history characteristics and strategies of the American eel, Anguilla rostrata. Can. J. Fish. Aquat. Sci 56(5):795-802.

Paulik, G.J. 1973. Studies of possible form of the stock and recruitment curve. Rapp. P.-v. Reun. Cons. Perm. Int. Explor. Mer 164:302-315.

Peterson, R.H. (Editor). 1997. The American eel in eastern Canada: stock status and management strategies. Proceedings of eel workshop, January 13-14, 1997, Québec City, QC. Can. Tech. Rep. Fish. Aquat. Sci. 2196: v + 174 p.

Porcher, J.-P. 1992. Les passes à anguilles. Bull. Français Pêche Piscicult. 326-327:5-14.

Richkus, W., and K. Whalen. 1999. American eel (Anguilla rostrata) scoping study: a literature and data review of the life history, stock status, population dynamics, and hydroelectric impacts. Electrical Power Research Institute, Palo Alto, CA. 1999. TR-111873.

Rosenberg, A., P. Mace, G. Thompson, G. Darcy, W. Clark, J. Collie, W. Gabriel, A. MacCall, R. Methot, J. Powers, V. Restrepo, T. Wainwright, L. Botsford, J. Hoenig, and K. Stokes. 1994. Scientific review of definitions of overfishing in U.S. Fishery Management Plan. NOAA Tech. Memor. NMFS-F/SPO-17. 45p.

Secor, D.H. and J. Rooker 2000. Is otolith strontium a useful scalar of life cycles in estuarine fishes? Fisheries Research 46:359-371.

Schmidt, J. 1922. The breeding places of eel. Phil. Trans. R. Soc. 211:179-208.

Smogor, R.A., P.L. Angermeier, and C.K. Gaylord. 1995. Distribution and abundance of American eels in Virginia streams: tests of null models across spatial scales. Trans. Amer. Fish. Soc. 124:789-803.

Taylor, A.H. 1995. North-south shifts of the Gulf Stream and their climatic connection with the abundance of zooplankton in the U.K. and surrounding seas. ICES J. Mar. Sci. 52:711-721.

Taylor, A.H. 1996. North-south shifts of the Gulf Stream: ocean-atmosphere interactions in the North Atlantic. Int. J. Climatol. 16:559-583.

Tesch, F.W. 1977. The eel. London: Chapman and Hall. 434 p.

Walters, C., and J.-J. Maguire. 1996. Lessons for stock assessment from the northern cod collapse. Reviews in Fish Biology and Fisheries 6:125-137.

Wenner, C.A. and J.A. Musick. 1974. Fecundity and gonad observations of the American eel, Anguilla rostrata, migrating from Chesapeake Bay, Virginia. J. Fish. Res. Bd. Can. 31:1387-1391.

Williams, G.C. , R.K. Keohn, and J.B. Mitton. 1973. Genetic differentiation without isolation in the Amercian eel, Anguilla rostrata. Evolution 27:192-204.

82 Williams, G.C., and R.K. Keohn. 1984. Population genetics of North American catadromous eel (Anguilla). pp. 529- 560 In Turner, B.J. (Ed.) Evol. Gen. Fish. New York: Plenum.

83 APPENDIX 3

Personal Communications

Durif, Caroline. Unité Ressources Aquatics Continentales, 50 avenue de Verdun, 33612 Cestas cedex, France.

Haro, Alex. S.O. Conte Anadromous Fish Research Center, Biological Resources Division, US Geological Survey, P.O. Box 796, Turners Falls, MA USA 01376.

Matthews, Milton. Erne Eel Enhancement Programme, Northern Regional Fisheries Board, Station Road, Ballyshannon, Co. Donegal, Ireland.

Therrien, Jean. Groupe-conseil Genivar, 5355 boul. des Gradins, Québec, QC, G2J 1C8 Canada.

Verdon, Richard. Hydro-Québec, Hydraulique et Environement, 855 Sainte-Catherine Est, Montreal, QC, H2L 4P5 Canada.

84 APPENDIX 4

Definition of Terms Used in the Document

American eel life history:

Catadromy–Life cycle in which spawning occurs in the ocean, feeding and growth occur in estuaries and fresh waters, and adults return to the ocean to spawn. American eels are catadromous, and they die after spawning once.

Panmixis–Single breeding population over the entire species range, without population substructure. American eels are panmictic.

Recruitment–Immigration of young stages (elvers, young yellow eels) to continental (coastaland in land) waters.

Leptocephalus–The oceanic, pelagic larval stage of eels. This stage lasts several months in the American eel.

Glass eel–Small, transparent eel formed by metamorphosis of leptocephali. Metamorphosis occurs at sea, perhaps near the edge of the continental shelf. Glass eels are unpigmented elvers. Glass eels enter coastal waters and estuaries and many ascend rivers during winter and spring.

Elver - Small juvenile eel. Term some time used vaguely, but morely refers to the first year in continental waters.

Yellow eel - Juvenile eel residing for feeding and growth in continental waters. The color of the specimen is not diagnostic of this life stage. This stage typically lasts several to many years.

Silver eel - Sexually maturing eel migrating to the oceanic spawning area. The color of the specimen is not diagnostic of this life stage. Metamorphosis to this stage includes changes in body color, structure and physiology of the swim bladder, and structure and physiology of the eye.

Other Definitions

Dip net - An active capture gear consisting of a rigid frame filled with netting, firmly attached to a rigid handle and manually operated by a single person.

Fyke net - A funnel-shaped net designed to intercept and setain moving aquatic organisms. The net is of varying length from cod end to wing tips and is fitted with various size netting. For glass eels the net measures 0.3 cm (1/8 in) mesh square measure or less.

Hoop net - A stationary cylindrical net fitted with mesh that is placed at the bottom of a body of water. The gear includes wings or leads attached to the mouth of the net.

Pot - A cylindrical or rectangular trap with funnels that is baited. The gear is typically made of mesh.

Sheldon eel trap - A box trap with netted wings used to intercept and capture glass eels and elvers.

Spear - Historically the most widely known and used method of capturing eel. Often consists of a spatula-shaped center piece with multiple teeth on each side, each tooth having a single barb. A 3-9 m (10-30 ft.) long wooden pole is attached to this instrument for probing the soft muddy bottom, sometimes through a hole in ice or from a boat.

Trap - Passive gear similar to, but smaller than weirs. May have one or two wings facing upstream to capture descending silver eel. Wings, if present, do not block entire stream and the unit is considered portable.

Weir - A trapping device consisting of two wings extending from opposite shores of the stream running obliquely downstream and converging to form a funnel, to which is attached a box trap. As silver eels descend streams, the wings guide them into the box trap. This passive capture gear is semi-permanent, constructed of wood or other solid material, and usually blocks most or all of the channel.

85 APPENDIX 5

List of Participants, Working Group on North American Eels, St. Andrews, N.B., Canada, 28 August – 1 September 2000

Name Address Telephone Fax E-mail Larry Marshall Dept. of Fisheries and Oceans +1 902 426 3605 +1 902 426 6814 [email protected] (Chair) Diadromous Fish Division fo-mpo.gc.ca BIO, 430 Polaris Bldg. P.O. Box 1006, Dartmouth, NS B3K 2A4 Canada Jacques Boubée National Institute of Water and +64 7 857 0174 +64 7 856 0151 [email protected] Atmospheric Research ri.nz P.O. Box 11-115, Hamilton New Zealand David Cairns Dept. of Fisheries and Oceans + 1 902 566 7825 +1 902 566 7948 Cairnsd@dfo- Diadromous Fish Section mpo. gc.ca Charlottetown, PEI C1A 7M8 Canada François Caron Service de la Faune Aquatique +1 418 521 3955 +1 418 646 6863 Francois.caron@f 675 Boul. René-Lévesque, Est ext. 4377 apaq. Québec, QC G1R 5V1 gouv.qc.ca Canada John Cassleman Science Development and +1 613 476 3287 +1 613 476 7131 john.casselman@ Transfer Br, Aquatic mnr. Ecosystems Science Section, Glenora Fisheries Station, R.R. gov.on.ca #4, Picton, ON K0K 2T0 Canada Martin Dept. Fisheries and Oceans +1 418 775 0634 +1 418 775 0740 Castonguaym@df Castonguay Institute Maurice-Lamontagne, o-mpo.gc.ca 850 Route de la Mer C.P. 1000, Mont Joli, QC G5H 3Z4 Canada Gerald Chaput Dept. of Fisheries and Oceans +1 506 851 2022 +1 506 851 2147 Chaputg@dfo- Diadromous Fish Section P.O. mpo. gc.ca Box 5030, Moncton, NB E1C 9B6 Canada Willem Dekker Netherlands Institute for +31 255 564 712 +31 255 564 644 [email protected]. Fisheries Research, nl P.O. Box 68, Haringkade 1, NL- 1970 AB IJmuiden, Netherlands Brian Jessop Dept. of Fisheries and Oceans +1 902 426 2158 +1 902 426 6814 [email protected] Diadromous Fish Division -mpo.gc.ca BIO, 430 Polaris Bldg. P.O. Box 1006, Dartmouth, NS B3K 2A4 Canada Alastair Mathers Ontario Ministry of Natural +1 613 476 8733 +1 613 476 7131 alastair.mathers@ Resources, Glenora Fisheries mnr. Station, gov.on.ca RR#4, Picton, ON K0K 2T0 Canada

86 Name Address Telephone Fax E-mail James McCleave School of Marine Sciences +1 207 581 4392 +1 207 581 4388 mccleave@maine. University of Maine, edu 5741 Libby Hall, Orono, ME 04469-5741, USA Dave Meerburg Dept. of Fisheries and Oceans +1 613 990 0286 +1 613 954 0807 Meerburd@dfo- 200 Kent Street mpo.gc.ca Ottawa, Ont. K1A 0E6 Canada Richard Peterson Department of Fisheries and +1 506 529 5929 +1 506 529 5862 [email protected] Oceans, fo-mpo.gc.ca St. Andrews Biological Station 531 Brandy Cove Rd., St. Andrews, NB E5B 2L9 Canada John Ritter Dept. of Fisheries and Oceans +1 902 426 3136 +1 902 426 6814 [email protected] Diadromous Fish Division o-mpo.gc.ca BIO, 430 Polaris Bldg. P.O. Box 1006, Dartmouth, NS B3K 2A4 Canada Robert Rosell Dept of Agriculture and Rural +44 1232 255 506 +44 1232 382 244 robert.rosell@dan Developmant for Northern i. Ireland gov.uk Newforge Lane, Belfast, BT9 5PX, Northern Ireland, UK David Secor University of Maryland Center +1 410 326 7229 +1 410 326 7318 [email protected] for Environmental Science, .edu Chesapeake Biological Laboratory, P.O. Box 38 Solomons, MD 20688 USA Henrik Sparholt ICES, Palaegade 2-4, +45 33 15 42 25 +45 33 93 42 15 [email protected] DK-1261, Copenhagen K ext 226 Denmark Peter Todd Ministry of Fisheries +64 3 5481069 +64 3 5469327 [email protected]. Private Bag 14, nz Nelson, New Zealand Victor Vecchio New York Dept. Environment +1 516 444 0476 +1 516 444 0434 [email protected] Conservation, Division of Fish, c. Wildlife and Marine Resources, state.ny.us 205 North Belle Mead Rd., East Setauket, NY 11733 USA Julie Weeder Maryland Department of +1 410 643 6785 +1 410 643 4136 mata-fish- Natural Resources, [email protected]. Matapeake Work Center, md.us 301 Marine Academy Dr., Stevensville, MD 21666 USA Hakan Fiskerriverket Kustlaboratoriet hakan.westerberg Westerberg Nya Varvet 31 @ 426 71 Vastra Frolunda, fiskeriverket.se Sweden

87